You are on page 1of 160


No. 24

Graphology and tke
Psychology oi Handwriting

Professor of Psijcholoflv,, Uulver.itij of Wyoming

William O. Owen
In recognition of his devotion

Truth as an Ideal

Introduction I

The Basal Concepts of Graphology 6
1. Handwriting as a form of Dramatic Expression.... 6
2. The Central Factor in Handwriting Individuality. . . 7
3. Control in Handwriting 10
1 1
4. Variability
5. The Theory of Signs; The Graphological Portrait... 13
Graphological Methods 17
1. The Method of Analogy 17
2. Appeal to Psychological Principles 18

3. Empirical Observation and Comparison 20
(Effect on writing of Education; Profession; Age;
Sex; Fatigue; Disease).
4. Intuitive Method 26
5. Experimental Graphology 29
6. Pathological Writing 35
The Graphological Elements (Comparative Survey of
Graphological, Mechanical, Pathological, and Ex-
perimental Investigations) 39
1. Graphic Dimensions / 39
2. Force of Movement; Pressure; Line-Quality 46
3. Direction of Movement 51
A. Slant 51
B. Alignment 57
4. Continuity of Movement 63
5. Proportion above and below base line 67


Disguised Handwriting 71
Intra-Individual Variability (Influence of Mental and
Physical condition on Size, Slant, and Align- /

Graphic Individuality, (Comparison of Handwriting Pat-
tern with characteristic Expressive Movements) ... 97
VIII. Graphological Study of Handwriting of Psychologists.
(Correlation of graphological and Characterolog-
ical 105
1. Preoccupation with details versus preoccupation with
principles in
2. Feeling of Self-Worth 113
3. Originality versus Organizing Capacity 116
4. Aggressiveness 1 19
5. Temperament 120
6. Explosive versus Inhibited Make-up 126
IX. Summary and Conclusions 132
References 139
The following studies are designed to canvass the possi-
bility of a scientific characterological utilization of hand-
writing. Their main purpose is one of orientation, prelim-
inary to an attempt to use graphic activity in tests of tem-
peramental or character traits, tests which are now in pro-
cess of standardization.
The discussion falls into two parts. Part I consists mainly
of a critical comparison of graphological contentions and the
outcome of modern scientific investigations of handwriting.
Part II reports a number of experimental studies, designed
largely to try out various methods of approach. Free use is
made of results from other experimental investigations by
myself which have been previously reported. The chapter
on "Disguised Handwriting" is reprinted, with a few minor
additions, from the Journal of Applied Psychology.
The "Graphological Study of the Handwriting of Psychol-
ogists" was made possible through the generous assistance
of those psychologists who furnished me with the material
necessary for a characterological rating. In the study on
"Intra-Individual Variability," I am greatly indebted to John
E. Anderson for faithful cooperation and in other studies I

have had the kind assistance of my pupils and colleagues at
the University of Wyoming. I take this opportunity of ex-
pressing to all who have aided me my heartiest thanks.
University of Wyoming, August, 1918.
Part I


The present day preoccupation with applied and, partic-
ularly,with vocational psychology has revived an interest in
attempts to analyze character by mearis of physical traits or
objective products. We have, for example, systems of
character analysis based upon so-called fundamental physi-
cal variables such as pigmentation, form, size, structure,
and expression. Graphology as an alleged science of psy-
chodiagnosis utilizes a particular form of expression, name-
/ly, handwriting. Graphology as so defined should, how-
ever, be discriminated from the graphology which is a study
of graphic signs of service in the identification of writing.
Both uses of the word are current today ; in our present dis-
cussion we are mainly interested in the former.
It has been assumed by many scientific workers that
graphology as a system of character diagnosis is on a level
with other pseudo-sciences which look for a facile interpre-
tation of one's mental make-up from a reading of the lines in
the palm of the hand or the bumps on the head. And, in
fact, certain extravagant claims of certain ambitious grapho-

logists relative to the possibility of determining the color of
a writer's eyes or the shape of his nose or the elasticity of his
bank-account from his chirographic style justify a healthy
incredulity. The best graphologists show, however, a cau-
tion and conservatism in interpretation that wins in a meas-
ure the reader's confidence and a desire to hear what those
of best repute have to say in defense of their art.
So far as the details of graphological diagnosis go one
may, indeed, be exceedingly sceptical and yet unwilling to
dismiss the whole matter on the ground that graphology is
on a par with palmistry, phrenology, or astrology. How-

ever mistaken and overly optimistic graphologists may have
been however obviously inadequate their control of observa-

tions, the precipitate from the extensive study of such men
as Preyer, Crepieux-Jamin, Meyer, Schneidemiihl, and
Klages certainly deserves respectful consideration. More-
over, on the general ground of intimate relationship between
consciousness and motor expression, the graphic pattern
which we call handwriting individuality demands careful
scrutiny. Possibly the determinants of handwriting indi-
viduality may be wholly external to the individual's psychic
make-up, but the statement should at least be accompanied
by an interrogation point. A more conservative position
grants the possibility at some future day of utilizing hand-
writing in psychodiagnosis but would defer such an attempt
until a more perfect technique has been acquired by psychol-

ogists for analysis of the grapho-motor process and the
graphic product. The wisdom of this position is evident
but an endeavor to try out immediately certain graphological
principles seems justifiable for the following reason.
Mental testing which is opening out into such tremen-
dous possibilities with reference to the analysis of the intelli-
gence make-up is baffled by the problem of diagnostic

tests for character and temperamental traits. But the need
of such tests is as obvious as difficulties in the way of get-
ting them are great. Especially necessary in vocational
selection is a determination of character qualifications. Two
individuals of equally keen intelligence may be very un-
equally fitted for the same position by reason of difference
in degree of persistence, energy, ambition, self-confidence
and the like. After a certain level of intelligence is attain-
ed, barring cases of exceptional ability, success in life would
seem to be dictated more largely by temperamental qualities
than by mentality status. The need for diagnostic tests of
such qualities therefore renders unnecessary any extensive
apology for excursions even into debatable territory. The
slightest chance of stumbling upon a useful suggestion of
procedure justifies such an excursion.

But only those readers who have attempted to work
through it are aware how voluminous is the literature of
graphology. This literature is found mainly in the French
and German languages. Discussions of graphology by
English writers suffer greatly by comparison, so much more
subtle, discriminating, and scholarly are the former. A
somewhat close reading of a number of these works has
convinced me that a summary of the points of view involved
might be profitably undertaken. If nothing more, it may
stimulate the psychology of handwriting to novel methods
of approach.
In my survey of graphological doctrine I have followed,
largely, a comparative treatment. I have sought, that is, to
present graphological material in light of the perspective
furnished by the modern scientific study of handwriting.
Such study has pursued several differing lines of interest,
very differently motivated. Aword as to each will serve
to map out our territory.
1. The most scientific study of
handwriting has centered
its interest in a delicate analysis of the graphic movements

as a form of motor reaction. Such investigators have de-
veloped and are perfecting an elaborate instrumental tech-
nique. One recalls in this connection the fine work of Gross,
Diehl, Meumann, Freeman and others.
2. Much less controlled and objective in its methods but
valuable as an attempt to approach the analysis from another
angle are attempts to determine certain mental factors that
condition handwriting the study, for example, of the imag-

inal and sensational control processes (Downey) or the fac-
tor of unconscious imitation (Starch).

3. Athird investigation, motivated primarily by peda-
gogical interests, has served to establish definite standards
for the evaluation of writing speed, quality and
Various scales of evaluation of graphic products have been
put to extensive use, among them the widely known scales
of Ayers and Thorndike.

4. Under pressure ofpractical need for accurate identi-
fication of handwriting for legal purposes, both civil and
criminal, the handwriting experts have given us detailed
analyses of the effect upon penmanship of writing systems
and writing apparatus. They have stimulated discussions
of the limits of variability and of disguise in the hands of
individuals and have adopted a procedural technique that
offers many a suggestion to the student of other aspects of
the general problem. Particularly stimulating is the ac-
count by Osborn of the utilization by the expert of the most
modern of mechanical appliances such as the document
microscope, the enlarged photograph, and delicate scales for
determination of line-width, degree of curvature of connect-
ing lines and similar graphic details.
5. Very greatly in contrast to the regulated analysis of
the legal expert are the descriptions of pathological writing
furnished us by psychiatrists. Yet pathological writing af-
fords, as we shall see, a unique method of checking up con-
clusions reached by other forms of procedure.
6. Lastly, we may list a line of investigation somewhat
difficult to characterize. It consists in the utilization of

writing as a material which may be employed in discovering
certain types of perceptual and judgmental reaction.
Strictly speaking our interest in this connection is not in
the psychology of writing, as such, but in the psychology of
the observer of graphic products. This latter aspect of the
situation is implicit in graphology in so far as the grapholo-
gist gives evidence of virtuosity in his handling of graphic
material. Binet's investigation of the graphologist ap-
proached the subject from this angle. Psychological inves-
tigation makes a similar approach in utilizing handwriting
as a material for study of the subjective judgment in trac-
ing family resemblances, and in training the expert judge of
handwriting for school systems. The insistence by Osborn
that extreme blindness to graphic form on the part of the
presiding judge may seriously handicap presentation of evi-
dence in the courts also stresses the significance of the sub-

jective factor in the applied psychology of handwriting.
Ourgeneral plan of procedure may now be outlined. The
first part of this little volume proposes to discuss, with the

critical background furnished by the specific investigations
listed above in addition to general psychological theory, the

following topics;
^ i. The general concepts upon which the graphologists
build their elaborate structure.
2. The methods they have utilized in their endeavor to
correlate particular graphic with particular mental or
temperamental traits.

. 3. The significance of certain graphic elements as de-
termined by graphology and by psychology.
The second part of the book will be devoted to reports on
experimental investigation of a few specific graphological
assumptions. Since the question of method of attack will,
necessarily, be in the foreground, meagre positive results
may be anticipated. Whatever positive conclusions are
suggested will encourage us in our search for diagnostic
material. The particular problems we shall attack in this
part include handwriting disguise as basal to all efforts to
discriminate between the spontaneous and the controlled
hand intra-individual variability
; graphic individuality
; ;

and, lastly, a graphological reading of a collection of hands
checked by a characterological rating obtained by a modi-
fied form of the order of merit method.

From my reading of graphological literature, five concepts
emerge as an understanding of the general as-
essential to

sumptions. These concepts may be listed as follows:
(i) Handwriting as a form of emotional or dramatic ex-
pression; (2) Graphic individuality as an outcome of cen-
tral factors; (3) The limits of voluntary control and its

significance for psychodiagnosis from written products ;

(4) The range and significance of graphic variability; and
(5) The conception of the graphological signs and of the
graphological portrait. Let us discuss briefly what is in-
volved in each of these concepts.
I. Handwriting as a form of Dramatic Expression.

A general presupposition of graphology is that all move-
ment has dramatic nuance and mirrors temperamental ten-
dencies. Writing is described as an expressive movement
on a par with gesture and emotional attitude. The French
characterization of writing as composed of "petits gestes"
conveys distinctly the point of view involved.
It may be said that this fundamental conception, even if

true, is too general to be of much value. A
serviceable ap-
plication demands a definite comprehension of the principles
underlying emotional and dramatic expression and its ap-
plication to a series of abbreviated movements that are pri-
marily constrained by the demands of social communication
to the production of stereotyped signs that may vary only
within certain prescribed limits. At best we have a baffling
entangling of external and internal factors. Training,
practice, convention "rigidify" the petit gesture. Yet, of
course, all emotional expression both yields to and yet
moulds conventional requirements, witness such forms as
tone of voice, facial accents, bodily postures. How far, in
fact, emotional attitudes are an outcome of conventionalized

and traditional expression, how far dominated by racially
ingrained patterns is itself a problem.
We shall have occasion to consider this question at closer

quarters when reviewing the graphic signs that are selected
as significant of emotional tendencies. Certainly, the broad
classification of outgoing movements (eccentric) and
movements of withdrawal (concentric) as characterizing
respectively attitudes of aggression and of defense cannot
sustain too great a weight. The situation will be canvassed
more specifically when we come to deal with variations in
Slant and Alignment, which are thought to carry emotional
2. The Central Factor in Handwriting Individuality.

The existence of graphic individuality, often of a very
pronounced type, will hardly be questioned. Its explanation
is, however, open to discussion. Is individuality in hand-

writing a product of objective factors only? Or in despite
of these does writing assume a specific character ?
Perhaps the cleanest-cut treatment of handwriting indi-
viduality, although not necessarily the most convincing, is

that furnished by the practical expert. It is of course his
contention that no two hands are ever precisely identical ;

he is willing to stake his professional reputation upon the
possibility of handwriting identification but he is apt to add
that the individuality with which he deals is an outcome of
the multiplicity of factors involved. Graphic individuality
is acquired ;
its origin is to be sought in the system learned

in school, in acquired habits of arm, wrist and finger move-
ment, in the kind of writing apparatus that is utilized, in the
amount of practice, in professional requirements, social imita-
tion and 'the like. On a low basis of calculation as to possi-
ble variations in writing characters which have been deter-
mined by careful analysis, Osborn tells us that "the mathe-
matical probability of two complete handwritings being iden-
tical is one in
something more than sixty-eight trillions."

The graphologists, on the other hand, assume that the
individual stamp of penmanship is a matter largely of cen-
tral "Handscript" is, essentially, "Hirnschrift."
As evidence of this they emphasize an observation frequently
found in the literature of the subject that the writing of a
given individual may be variously produced so far as mus-
culature is concerned and still bear the individual stamp.
We are told specifically that thefoot-writing and the mouth-
writing of a given individual resemble his hand-writing.
(39:37) It is, moreover, a matter of general observation
that one's free arm, magnified writing on the blackboard re-
sembles one's finer pen-script executed perhaps with delicate
finger movement.
The presence of central factors could not, of course, be
cited as decisive so far as significance of graphic individu-
ality is Objective conditions might leave a de-
posit of memory images of design and proportion of letters
and the like which might well function in spite of shift in
peripheral musculature or writing apparatus. But, obvious-
ly, the graphologist must mean something more than this.
His conception of graphic style carries implications of the
same stamp being impressed upon other movements execut-
ed by the individual in which there could be no question of
the operation of specific graphic habits. He has, indeed,
claimed that a similarity exists between a man's handwrit-
ing and the manner in which he walks or gesticulates he has

asserted that the pencraft of the painter mirrors the peculiar
distinction that marks the sweep of his brush across the can-
I knowof no controlled observations supporting such
statements. Obviously there must be a wide range for error
if such conclusions are based merely upon casual observa-

tions that are motivated by definite expectation and interest.
In the second part of the book I shall report an attempt to
test the assertion of the existence of individual motor pat-
terns which stamp gesture and walk and posture as well as

Psychologists who have made a special study of writing
movements have had little to say of graphic individuality.
Hirt(23a:386) speaks of unknown biological laws which
are basal to handwriting individuality. Meumann holds, "It
is the na,ture of innervations coming from the cerebral cor-

form and extent to which they are synthesized, that
tex, the
is main determinant of the character of the writing."
Movement-individuality is probably a product of many
different faqtors more or less fundamental to personality as
a whole. Such, for example, are sensory predispositions,
motor degree of unidextrality and the like. Many of
these contributing factors might well be discussed in detail
but since individuality is a function of the complex as a
whole rather than that of the elements as elements we may
postpone certain observations until later. Series of samples
showing the genetic development of hands would be of very
great value in helping us to analyze the appearance and the
consistency of graphic individuality. The great individual
variation in the time at which handwriting sets or matures
with a consequent fixation of "style" is one of the most in-
teresting aspects of writing with which I am acquainted.
Time of fixation appears in many instances to be a family
trait and opens up the question of the possibility of innate
and very fundamental tendencies reflected in the existence
of family hands unless, of course, the family type be wholly
the product of social and educational environment. Person-
ally, I am convinced that family resemblance in chirography
is not to be
explained on the basis of similarity in teaching
and social models. Nor indeed do I find its most impressive
aspect in similarity in graphic details such as design of let-
ters but rather in the appearance of
general motor patterns,
a fluidity or rigidity of movement, an inflection of manner
that seems to lie back of the assumed features. Hirt's ob-
servations on this point are of particular value and his in-
teresting citation of his own case carries considerable weight
in favor of a heritable factor
being involved in the situation

(23a:386). Myown collection of cases carries the same
import. (14b) While Thorndike on the assumption of a
native factor has utilized the resemblance between the hand-
writing of twins as a means of getting a scale of unintention-
al resemblance. (46b) More recently still Starch has re-
ported a correlation of .72 for speed of writing in a group
of eighteen pairs of adult siblings, and a correlation of .46
for quality of writing. (43b)
3. Control in Handwriting.
In a sense the concept of voluntary control is the crucial
one in graphology. Graphologists in accepting a specimen
demand that it be a "bona fide" article. They refuse to
handle what they call calligraphic or purely conventional
hands hands that lack individuality. They would seem to
recognize the possibility of writing from which all individu-
ality has been squeezed by pressure of professional necessity
or by need for disguise. What it sought is a handwriting
specimen in which the individual gives way to natural im-
pulses of expression. They prefer for their purposes the
free writing which one addresses to one's self in rapid note-
taking or the writing in informal letters to intimates rather
than the chirography on stilts which one assumes to impress
another or for formal examination purposes.
That the limit of control is the crucial point in handwrit-
ing identification has been clearly seen by the legal experts.
Schneikert (41), for example, in planning for the Berlin
police system a method of classification of the handwriting
of criminals for purposes of identification bases his general
scheme on the elements in writing that may be easily dis-
guised or the reverse. Such graphic characteristics as size,
slant, form of letters are easily modified at will while rela-

tive proportion between one and two or three space letters,

continuity of writing, mannerisms in dotting the i, etc., re-
main relatively constant even in an attempt to disguise writ-
In general, the success of the legal expert depends upon
his knowledge of where to examine writing with the expec-

tation of seeing the mask dropped. The most careful dis-
guise through at some point. Some trick in mak-
lets light

ing a comma, or crossing the "t" may give away the secret.
Furthermore, only the expert is aware of the significance
of what we may call graphic inconsistencies, the appearance,
for instance, of an alien letter-form in a writing of a partic-
ular type. Almost supernormal control is involved in inhi-
bition of the numerous habits that have been evolved in the
course of learning to write. Moreover, this excessive con-
trol manifests itself in the appearance in such artificial writ-

ing of numerous evidences of writing inhibition, hesitations,
and retouching of letters. Often the flowing writing move-
ment is replaced by a slow drawing movement which in itself

is indicative of a high degree of voluntary control or effort.

So much we learn from the legal expert. The experi-
mental graphologist has not been slow to utilize this con-
ception of control in his study of writing-types.
Klages (26b) discussing the meaning and limitations of
the concept of handwriting as a type of individual behavior
states that one should attempt to diagnose character from

writing only after a thoroughgoing effort to classify a par-
ticular writing with reference to the amount of control ex-
ercised in the writing-act. He classifies writing as either ( I )
artificial or (2) natural. Artificial writing includes dis-

guised, calligraphic, and ornamental writing. Under nat-
ural or spontaneous writing we get (a) a more controlled and
(b) a more involuntary type of writing. Control in writ-
ing may arise either from mastery of impulse or from exces-
sive inhibition. Involuntary or uncontrolled writing also
shows variations dependent upon acquired traits. It is

shown, however, that the concept of an acquired handwrit-
ing absolutely inexpressive of the writer's organization is
but a limiting notion. Power of disguise or a high degree
of sustained control are themselves significant traits.
4. Variability.
The concept of variability is allied to that of control. It

may be considered from two standpoints, first, that of spec-

ific variability in the writing of different persons under set
conditions ; and, secondly, range of variability in the writing
of a particular individual.
The extent and quality of graphic variations that may be
anticipated under given conditions such as emotional dis-
turbance, nervous disease and the like or those that are de-
pendent upon age, sex, and profession will be rehearsed in
other connections. We
may confine ourselves here to con-
sideration of variability in the writing of a given individual.
Is this variability so great as to prohibit all utilization of

writing in character-study ?
Absolute invariability in graphic products is, of course,
unthinkable. One of the most suspicious signs of a forgery
by tracery consists in an exact reproduction of a signature.
The existence of two absolutely identical natural autographs
is an impossibility, the experts tell us. Their comparisons
involve of course microscopic measurements and not the
mere testimony of the bare eye. But granting a variable
element in all graphic expression, is it so extreme as to lead
us to conclude that writing individuality is too fluid a thing
to have diagnostic significance?
Such a question could be answered only by an estimation
of the actual extent of variability found in the writing of a
given reagent under cited conditions.. In the hope of get-
ting some exact determination of the range of variability I
gathered the material and made the measurements reported
in one of the studies in Part II. The range of variability
was found to be pretty extreme but without loss of individ-
uality. Variability in particular graphic signs and the in-
terpretation of the significance of such specific variation is
another matter.
In any case, the common exclamation of laymen, "I never
write twice alike," is subject to big discount. The similarity
in writings that may have elicted such a comment one I
have heard again and again may be so striking as to lead
the experimenter to wonder at the blindness of the person in
question. Undoubtedly, however, one is more sensitive to

minor variations in one's own writing or in that of members
of one's household than to variation in that of acquaintances,
just as most of us are less ready than strangers to see re-
semblances in the family circle.
It should be noted at this point that the writing-expert is
accustomed to find graphic variability limited by the writing
habits which are a product of the kind of movement and
writing-systems learned and the amount of practice or
graphic expertness. Variability is introduced by objective
factors such, for example, as the haste or leisure with which
one writes, the quality of pen, paper, and ink, the innumera-
ble chance factors that have a casual and not a causal sig-
The best graphologists are aware of such factors condi-
tioning the appearance of a handwriting under examination.
They demand many specimens of a given writing produced
under varying conditions.. They refuse, frequently, to at-
tempt interpretation of unaccustomed, official, clerical, or
so-called calligraphic writing. They ask for running writ-
ing produced under natural conditions of interest in the
5. The Theory of Signs; The Graphological Portrait.
In organizing their systems and presenting a technique of
procedure, graphologists are wont to list so-called grapho-
logical signs or elements together with an interpretation of
their significance. Variations occur, of course, in the vari-
ous presentations. Under some form, however, all writers
upon the subject deal with such graphic elements as size,
alignment, slant, degree of continuity, proportion and the
like. Usually a multiplicity of causes for the same effect is
The mechanical aspect of much of the work with "signs"
has been definitely criticized and in practice the grapholo-
gists modify their analytic procedure by an attempt to in-
terpret each detail in the light of the setting in which it oc-
curs. In the conception of the graphological portrait, the
synthetic function of graphology is stressed at expense of

the analytic. The real measure of a graphologist's expert-
ness consists in his ability to interpret graphic signs in their
relation to one another and to the whole complex in which
they occur. He must synthesize a multiplicity of details,
reconcile opposing symptoms and succeed in locating the
central characteristic that furnishes the key to an under-
standing of the organization of diverse traits into a unique

A balancing process somewhat similar is urged by Dr.
Blackford (5) upon those who would use successfully her
system of character-analysis. It is not sufficient to list for
any individual the fundamental physical variables, color,
size, texture, form and the like. One must be skilled in de-
termining how one characteristic modifies another in the
way of accentuation or neutralization ; as, for instance, the

degree to which convexity of form annuls brunette coloring.
One must discover a unity back of apparent contradictions.
At this point scientific analysis is abandoned for artistic
creation. Just as little as a cataloguing of psycho-physical
traits can give us the living personalities of fiction and drama
can a tabular summary of graphic characteristics and their
significance give us the graphological portrait.
Two very interesting problems are involved in this con
ception. First, to what extent, if any, will science ever suc-
ceed in capturing the inmost citadel of personality? And,
secondly, can it force its way in by any such a tour de force
as that of the trained intuitions of the graphologist or of the
physiognomist or other student of expression?
The attempt of the psychologist to penetrate the secrets
of character-organization is evident in his recent torturing
of the instinctive life in his search for an all-sufficient prin-
ciple of interpretation. That he has made valuable discov-
eries must be conceded so far as his skill in twitching out
of the pattern certain threads is in question. His success
at synthesis is less evident as shown by the sense of violated
personality that his analyses leave with us in contrast to our
acquiescence in the portrayal of the same individuality by

the great dramatist or fictionist. Quite probably it is a
mistake to confuse the functions of science and art. So far
as theoretical psychology is concerned her function may well
be limited to furnishing a technique for character-analysis
with, at most, a cataloguing of certain temperamental and
character patterns and correlational formulae. But applied
psychology will probably not content itself with bare for-
mulae; it will develop the expert diagnostician whose ad-
vise anent matters of vocational pursuits or mental hygiene
will be very definitely controlled by the results of a clinical
examination conducted by himself or his specialists but will
involve further a fusion in the white light of character-
divination. The expert medical diagnostician also posses-
ses this synthetic activity to a high degree we are told. How
far such a gift is the outcome of original genius, how far
the result of previous analysis and rich experience it is im-
possible to say in our present ignorance. It brings us face
to face with many curious problems that are inherent in
the concepts of "Intuition" and "Automatic Acitivity" and
with certain aspects of the analytic versus the synthetic
method of handling material to which we shall return in a
later discussion.
The answer to our second question, the adequacy of cer-
tain forms of comprehensive character diagnosis, is very
simple. They have not as yet been able to substantiate
their claims. We hear of skilled readers of the human
countenance but where may one find reported a controlled
test of their actual proficiency? Business psychology is,
however, envisaging this problem at close quarters and it
may discover in time a specialized ability for speedy and ac-
curate estimate of human nature on the bases of physical
form and expression but from present indications it appears
more likely to recommend instead a complete substitution of
standardized mental tests as more accurate than the immed-
iate judgment of the most experienced judge of human

The proficiency of professional graphologists in delinea-
tion of the graphological portrait can be estimated only by
controlled tests. It is not easy to win the consent of a pro-
fessional to the precautions that are necessary. Binet (3c),
however, succeeded in gaining the confidence of the French
graphologists but in order to estimate their success or fail-
ure in numerical form he was compelled to limit his test to
investigation of more or less mutilated aspects of character-
reading, namely the determination of sex, age, and com-
parative intelligence and morality from chirography. The
specific results of Binet's investigations we shall have occa-
sion to refer to as the discussion develops. Suffice it to say
here that Binet concluded that although there was some-
thing of truth in graphology as practiced by his collabora-
tors, 'there was also much gross error and uncertainty.
Graphology he thought might, however, be a science of the
future. Quite possibly handwriting may be utilized in char-
acter diagnosis but in a much more modest form than is
implied in the notion of the graphological portrait.

In connection with a detailed discussion of the correlation
of mental traits with specific graphic signs, we shall have
occasion to handle this topic with some care. In this chap-
ter let us confine ourselves to general observations.
In reading graphological literature I have frequently
asked myself what evidence could be cited for particular
conclusions and, frequently, I have been unable to put my
finger on the method by which the conclusion was reached.
It is, in fact, a rather difficult matter classifying or charac-

terizing the kinds of evidence utilized. They are implicit
rather than explicit in the graphological treatment, as is
usually the case when a complex material is handled by a
method of procedure that may be described as intuitive
rather than analytical, artistic rather than scientific. An ex-
haustive study of the logic of graphology would probably
repay the investigator. I shall, however, content myself
with characterization of what I take to be the chief methods
utilized by graphologists, namely, (i)) The Method of

Analogy; (2) The Deduction of certain conclusions
from general Psychological 'Principles; (3) The Induc-
tive Method of Empirical Observation and of Compara-
tive Study of Group and Individual Variation; (4) The In-
tuitional Procedure; (5) Experimental Graphology; (6)
Pathological Writing.
In actual procedure these different methods are pretty
thoroughly intertwined and some violence is done to a par-
ticular treatment by twitching them apart. Nevertheless,
some schematism is necessary.
In "Breaking ground in every line of human endeavor re-
course to analogy is evident. In primitive mental life we
find rich material for study of the way in which
many cur-

ious practices may grow up through associations by simi-
larity. Wells, (note) quoting Josiah Moses, cites the fol-
lowing examples. "Bloodroot, on account of its red juice,
is good for the blood liverwort, having a leaf like the liver,
cures disease of the liver etc. These early analogies im-
press us as crude in the extreme and amusing as well as in-
consequential. Yet organization of experience is to a great
degree dependent upon fusion of experiences on the basis of
rough and, at first, superficial likenesses. And in more re-
fined forms this method continues to dominate thought.
My general impression from reading graphological lit-
erature was that considerable appeal was made to very sup-
erficial analogies. But I do not find that I can list many
such associations. I have the feeling, however, that the at-
tribution of a given significance to a certain graphic sign
was often based originally on a facile use of association by \

similiarity and that the psychological grounding is an after-
thought. Purely analogical, I should say, is the deduction
of the serpent-temperament that of the diplomat ( !) from
sinuous alignment or undulant bar of the t; and the paral-
leling of graphicand mental continuity, so that the quality of
coherent thought is ascribed to the writer of a connected
hand, and intermittent flashes of inspiration credited to him
who indulges in frequent breaks in graphic connection.
Analogical reasoning would seem to be involved in citing
illegibility as an indication of dissimulation, and the produc-
tion of small writing as an evidence of love of minutiae. A
similar kind of induction is apparently involved in the rela-
tion assumed between an extensive movement of the pen-
point above the line-heavenward as it were! and idealistic
proclivities, and ascription of earthly or materialistic quali-
t es to the penman producing long down-strokes.

Note:Mental Adjustments, p. 93.
2. Appeal to Psychological Principles.
The literature of graphology makes frequent appeal to
psychological principles. I have accordingly been some-

what surprised to find that my reading has precipitated so

little in the way of definite reference of graphological inter-
pretations to the specific psychology of movement. A few
citations are made to bear the weight of an elaborate super-
a. The general tendency of every psychic state to issue in
some form of movement is frequently cited as fundamental
to the graphological position although a limitation to its
serviceability as a principle of interpretation may be recog-
nized not only in its extreme generality but also in the dif-
ficulty in application due to individual differences in the ex-
pressive threshold.
b. A second Jaw appealed to is that of dynamogeny,
namely, that .the force and energy of movement are a direct
outcome of mental energy, which is, in turn, conditioned to
a degree by external factors such as external illumination,
temperature, writing apparatus, etc. Size of writing and
pressure are two graphic elements interpretation of which
is referred to this general principle. The magnified writing
of the ambitious and the minute penmanship of the tired or
cautious person are described as well as the heavy stroke
of the strong-willed and the light stroke of the weak-willed
c. A "third
principle is the so-called law of emotional ex-
pression which correlates centrifugal and centripetal move-
ment with joyful and depressed moods respectively. The
treatment is assimilated to much that we are familiar with
in other applications of emotional expression, for instance,
the Delsartean system of eccentric and concentric postures.
Granting that the general assumption of emotional expres-
sion is well grounded, we find some very dubious applications
of it in graphology. Mainly such application is to be found
in dealing with the significance of slant and alignment. Thus

falling alignment is a symptom of depression like the drag-
ging footstep rising alignment is indicative of hopefulness.

Back-slant is symptomatic of reserve just as extreme right
slant betokens emotional susceptibility and vertical writing
indicates emotional control.

Two other principles of interpretation are cited by Klages.
(26a). One (d) he calls the law of_periodic fluctuation
of attention which results in relaxation of control at cer-
tain parts of lines, words, and letters and hence makes it
possible to discriminate between voluntary and spontane-
ous graphic traits. Klages' chief contribution, however,
concerns (e) the operation of impulsion and inhibition in
determining certain graphic patterns. The application of
such a psychological principle does not favor the mechani-
cal listing of isolated graphic traits with their respective sig-
nificance so much as it does the discovery of the possible
combination of specific symptoms and their common refer-
ence to the general motor organization of the individual.
Klages also cites in explanation of certain graphic signs
a law of so-called feeling for spatial analogy; the implica-
tion being, I take it, that in certain cases a delicate applica-
tion of Empathy operates in determining preference for
certain forms and slants.

3. Empirical Observation and Comparative Study.
The empirical method consists in extensive study and
comparison of handwritings. Thus the hands of persons
of known common traits are compared and a conclusion
drawn as to the way in which a given psychic character man-
ifests itself in script. Or a certain grouping of hands re-
sults from one's study and may lead to a cataloguing of the
mental traits possessed by the penman in order to determine
whether there are any that are common to the group.
Scholarly graphologists amass collections of writing speci-
mens whidh may be utilized in the twofold way mentioned
above. Thus they may institute comparison between the
hand-writing specimens found in collections coming from
the intellectually inferior and the intellectually superior and
seek the graphic symptoms of intellectual superiority or in-
feriority. They may compare the chirography of criminals
with that of moral reformers and so on. This method may
be refined to any degree, with an application of the classic
method of agreement and difference.

In any case a criticism urged by Mr. Osborn is undoubt-
edly valid, namely, that too little information is current
among graphologists as to the general effect upon script of
the writing system learned. These systems vary not only
from one nation to another but within a given country from
one decade to another. Thus in the United States alone it

is possible to trace the vogue of at least five systems ; name-
ly, the old English round hand the modified round
hand ;

the Spencerian system ;
the modern vertical ; and, in addi-
tion, an angular style taught in schools for girls. Differ-
ence in designs of letters, difference in proportions between
letter-parts, difference in slant and shading will character-
ize the script of those taught different systems. Moreover,
mannerisms from foreign systems may cling to a style thus ;

the influence of German script on writing is a very percepti-
ble one. The unsophisticated observer may find striking
similarities and differences in two writings that have no sig-
nificance whatever other than witnessing similarity or dif-
ference in the system of writing that was learned originally.
As an outcome of this masking of the individual chirography
by national and epochal habits it would seem that we must
seek for the distinctive graphic sign of say diplomacy or
candor or imagination in penmen who have learned differ-
ent systems of writing or we may institute a comparison of
hands for determination of a particular difference only when
confident that in general the penmen have learned the same
system of writing.
But there are other factors which might influence the
grouping of a collection of samples but which have only a
limited psychodiagnostic significance. The best grapholo-
aware of these contributing elements
gists are perfectly well
and have given us a more or less detailed treatment of them.
W may list these factors as follows education and amount

of practice in handwriting professional requirements which

determine the vertical hand of the librarian, or the print-
like hand of the engineer age sex
; ; ; fatigue ;
and disease.
A word as to the significance of each.

number of observations point to a fairly general rec-
ognition of certain differences between the writing of the
educated and the uncultivated. The chirography of the lat-
ter shows an unaccustomedness, an awkwardness, an inex-
pertness that marks it rather unmistakeably. But one can-
not always apply the contrasting adjectives to the hand of
the educated except in so far as accustomedness is concern-
ed. It too may be both awkward and without grace, al-
though usually rapid. The fundamental difference grows
out of amount of practice in using the pen. Practice is evi-
dent in the greater speed and smoothness with which the
graphic product is produced. There exists, however, a so-
called cultured hand which indicates breeding to a very high
degree. It possesses both grace and distinction as well as
the facility that is the result of much use of the pen. With
the increasing utilization of the typist such hands are be-
coming rarer and the educated man is satisfied with a scrawl
as a mark of identification.
Since practice is so great a factor in development of
graphic virtuosity, we are not surprised to find certain lines
of work leaving an imprint upon chirography. The teach-
er's hand is conventional. The clerical hand is marked by
ease of manner, speed and greater or less conventionality.
Even more conventional, very deliberate and slow is the ver-
tical hand of the cataloguer. The telegrapher's hand is
rapid, fluent, marked by a definite style and exhibiting cer-
tain mannerisms as to the number of words per line and the
like. Such hands are often cited in support of the position
that handwriting individuality is the product of objective
factors only. Graphic virtuosity, with its accompanying
speed and satisfactoriness of outcome, is the result of cor-
rect and prolonged practice. All of us might, it is assum-
ed, acquire the rapid business hand or the artistic print of
the mechanic. The assumption is a big one. As have been
urged before, hands that show to a very high degree the pos-
sibility of voluntary control undoubtedly exist, but not every
one can produce them. Not every one can acquire as a per-

manent possession the smooth, rapid, highly legible clerical
or business hand. A
process of selection goes on so that
the individual who finds himself limited on the side of graph-
ic facility drops out of the race early. As the reverse to this
picture we may point out the existence of families of clerks,
one of whose assets is a rapid legible hand, pleasing
in appearance. I could cite many instances in the commun-

ity with which I am best acquainted where the easy acquisi-
tion of a ready handwriting has been noted for many mem-
bers of the family, and for two or three generations.
In one case this graphic ease has determined the line of work
adopted by many of the group.
Age _
is also _a very definite factor conditioning handwrit-
ing. In part, of course, it is a question again of amount of
practice. With increasing age conies increased motor skill.
In part, it is a question of varying degree of dominance of
a hand by the conventional standards of the system learned.
The young hand is less individual. In the decline that comes
with old age indicative signs appear in the writing. Loss
of motor control may 'be. manifested by tremor often writ-

ing is increased in size, because of failing eye-sight or as a
compensation for ataxia; there is an approximation to the
so-called masculine type.
Very real as these symptoms are, the deduction of age of
penman from a given handwriting is by no means easy.
Binet in his test of professional graphologists found that age
could be determined with some degree of accuracy, on the
average within about ten years (3c) but that there were
many specimens that gave little indication of age or even gave
a false indication. I have already spoken of the curious dif-
ference in the time at which writing sets, as a family char-
acteristic. In the case of a late maturing or setting of hand-
writing we may get the impression of immature or child-
like hands from specimens produced by well-developed indi-
viduals. I have long been curious as to the explanation for
this late setting of handwriting. Instances that have par-
ticularly attracted my attention include three cases of very

brilliant young men, gifted in law or literature, whose in-
tellectualdevelopment has been remarkably precocious but
whose hands have "ripened" very slowly. I am inclined to
think that a late maturing of a hand is an indication of a
"sensory" make-up; an early maturing of a "motor" make-
up, a distinction to which we shall return later.
Sex as a determinant of handwriting has been dealt with
at some length by investigators. Mostly, graphologists are
somewhat conservative concerning the possibility of de-
ducing sex from handwriting. Binet's test of graph-
ologists showed that their successes in detecting sex from
handwriting ranged from 63 to 78.8 per cent, and under
favorable conditions might reach 90 per cent (3c), figures
which were confirmed by a later experiment by myself.
(14c). Curious inversion of sex-signs were, however,
discovered by experimentation so that many women are
found to write hands that are unanimously chosen as mas-
culine hands and a few men write unmistakeably ladylike
hands. The interpretation of the situation is somewhat am-
biguous, since sex in writing may be largely an outcome of
social factors which emphasize neatness, grace, convention-

ality in the woman's hand and speed, force, and originality
in the man's. On the other hand, Meumann (32) reports
a masculine and a feminine type of hand, differentiated by
the appearance of characteristic pressure curves, types which
if confirmed would evidence rather fundamental differences

in kind of motor control.
effect of fatigue and of disease upon writing opens

up an extensive field for exploration. Professor Janet, of
the College de France, urges that experimental graphology
should begin with studies in pathological graphology, stud-
ies on the effect upon handwriting of diseases of motility
and sensibility, or of specific diseases, such as those of res-
piration and of circulation. From the more pronounced
modifications of handwriting transitions may then be made
to its more delicate inflections. This recourse to pathology
bids fair to prove increasingly fruitful and deserves treat-
ment in a separate section.

Within these generic types as outlined above, the graph-
ologistmust conduct his search for character-complexes,
guarding always not only against confusion of the generic
with the individual but also against the accidental variations
that are due to purely objective factors such as writing ap-
It is
paratus, illumination, haste, social requirements, etc.
the complexity of the problem that leads many psychologists
to question the possibility of a serviceable psychodiagnosis
from handwriting. They may grant the revelatory character
of movement and yet despair of any very specific utilization
of it so far as writing is concerned. Yet precipitates from the
extensive study and comparison of handwriting specimens
by able observers certainly deserve consideration. If noth-
ing else, their conclusions afford suggestions for an experi-
mental program.
If I may judge from my own experience, observation of

handwriting results in what may be described as conceptual
precipitates, composite images, very similar to the generic
images of facial types. I .find myself mentally classifying
hands as belonging to the "smooth flowing" type or the
"labored inhibited" type or the "rapid, explosive" ty^e.
Such categories have developed from my experience quite
without deliberation, although I find that I possess fairly
clear-cut visual images of one or two specific hands that
may stand as representative of a particular type. My men-
tal fixation of a hand involves its classification with a par-
ticular group. My failure to note details is surprising but
in spite of it or perhaps because of it I have a rather unus-
ual capacity for recognition and memory of hands. My
composite images of hands have developed in connection
with study of individuals so that as part of my classification
I am apt to question myself as to whether the penman in

question belongs temperamentally with the group in which
his writing places him. Thus the "flowing" type of hand
has so often in my experience been produced by the socially
tactful, adjustable, often charming, sometimes merely plaus-
ible, individual that I find it creates a definite expectation.

In addition to these "generic images" of hands that have
resulted in part, from a native interest in graphic expres-
sion, and, in part, from extensive preoccupation with chir-
ography as a material for experimental work, I find a sec-
ond factor involved in my sensing a personality from hand-
writing. This second factor may be described as an attempt,
although not an overt one, at motor mimicry or imitative
interpretation. I find myself imaging kinesthetically the

type of movement suggested by a given handwriting. The
imitation results in a feeling of the assumption of a foreign
personality. The general method is similar to an attempt to
get a clue concerning the permanent temper or casual mood
of another by mimicry of his walk or his attitude. Whoever
has tried copying another's carriage, his manner, for ex-
ample, of carrying his crooked arms with elbows outspread
in true Irish fashion or hugging the body in diffidence is
aware how enlightening such mimicry may be. The sup-
pressed mimicry of a graphic pattern is, of course, a much
more subtle matter. In my own case it is released only by
very individual hands and only when I am in certain frames
of mind. I have, however, seen such a method utilized in
very explicit form by a little girl of ten years who once served
me as subject. This child had recourse in the most naive
way to mimicry by facial expression and bodily contortion
of the handwriting she was observing.
4. The Method of Intuition.
In the section on the concept of the graphological portrait
we saw that there is a point at which graphologists aban-
don an analytical for a synthetic method of procedure. In
dealing with handwriting as material for study I am, there-
fore, tempted to discuss a little more fully the opposition be-
tween a deliberate and an intuitional handling of material.
This discussion is motivated by two observations, first, the
extraordinary differences that normal individuals show in
their capacity to recognize and remember
and, secondly, by the distrust of the handwriting expert of
judgments based on general appearance of writing.

Binet in his investigation of the extent to which age, sex,
intelligence, and morality could be told from script found
that such measure of success as was achieved by profession-
als could be approximated by amateurs. My experience
with reagents in an experiment on the sex judgment result-
ed similarly. (14c). R, in particular a highly sensitive
young woman with decided literary and artistic gifts gave
evidence of extraordinary facility. Her record is very
nearly as accurate as that of Crepieux-Jamin, the French ex-
pert. R. reported a very definite sensing of personality
from writing, the accuracy of which could not be tested as
her help was available only for the one series of experi-
In a number of other experiments in which handwriting
has been used as material a great individual variation in
facility in handling it is evident, quite apart from training
or extensive experience with it. In an experiment in which
reagents matched pairs of addresses written by a given num-
ber of penmen, I found not only a wide range of variation

in abilitybut also an approximation of the best record by a
girl of eleven years. In an experiment on disguised hand-
writing I found one wholly unpracticed reagent whose pene-
tration of a disguise excelled that of very careful students of
the subject. Osborn found a practical application for such
individual difference in its negative aspect; failure to see
similarity in handwriting often makes it impossible for a
judge to follow the line of argument presented by a hand-
writing expert in court. The Osborn test for determination
of what he calls form-blindness namely the search for
samples of words written by the same penmen revealed a
wide range of variation in individual records, namely, from
100 per cent accuracy in 8 minutes and 35 seconds to 65 per
cent of accuracy in 9 minutes and 55 seconds." (36:6)
Bingham (4) reports, in comparison with other tests, a very
high coefficient of variability for the Osborn test.
That the amateur should so closely approximate the rec-
ord of the professional points to an interesting problem, if

only the identification of a field of work in which practice
effects are at a minimum. It raises the question, for a spec-
ific situation, of the value of the intuitional method. Dear-
born (lib) has recently analyzed in most profitable fash-
ion what he conceives to be involved in intuition and he has
urged the psychologist to enter upon a scientific investiga-
tion of this very promising concept. Rather than approach
the problem as a variation in sex-intelligence, as Dearborn
suggests, it would seem more auspicious to map out the field
on the basis of different materials. Specialized virtuosity in
any field and, in particular, the automatic processes of art
would suggest themselves as promising material for analy-
sis. The contributions of both original capacity and exten-
sive training could possibly be laid bare.
In myexperience with reagents in tests on handwriting I
have noted two varying tendencies ;
one a preoccupation
with graphic details, the other a preoccupation with the gen-
eral appearance of the hand in question. I cannot say on
the basis of the results which is generally the more success-
ful; but there can be no question that in some instances a
preoccupation with details has completely blinded a reagent
to very striking individuality. It is with considerable aston-
ishment that I have observed the insensitiveness to general
appearance of certain very careful and highly intelligent
reagents who compare varying details with most painful
exactness and yet totally miss the graphic pattern. This
point I have discussed elsewhere (14,d) but only in such a
way as to set the problem. Possibly we have here an in-
stance of judgment of general likeness (impressionistic)
versus one based on specific difference.
The problem involved is very extensive in its application.
Work in systematic botany and zoology reveals, I amtold,
the same sort of distinction in scientists.Atoo-great pre-
occupation with similarities may lead to an oversight of dif-
ferences that may prove basal from a classificatory stand-
point, while preoccupation with differences may result in the
endless splitting of subdivisions. In an experiment on

handwriting similarity and difference I sought to determine
whether there existed an individual difference in the readi-
ness with which difference or similarity was perceived. The
results seemed to indicate that most reagents are able to
shift somewhat easily from one mental set to another but
that there were reagents who were actually more successful
in maintaining one or the other of the two sets. The most
striking example of facility with likeness and incapacity to
handle differences occurred in the case of a girl whose fail-
ure in botanical classifications was in great contrast to her
usual academic success.
The handwriting expert in court procedure is most dis-
trustful of ajudgment based on general appearance of hand-
writing. In study of handwriting he recommends placing
word by word and letter by letter the material from the dis-

puted document and the possible original. His judgment
is the outcome of the most refined measurement and com-

parison of details. There can be no question of the just-
ness of the expert's insistance upon the methodical and ex-
acting testing of a questioned document nor his scepticism
of the unchecked and biased testimony of the average un-
discriminating witness. But it would, none the less, be of
great value to institute a comparison under controlled
conditions of the judgment of a picked reagent based on
general appearance and that of the expert based on a com-
parison of details. The latter method makes possible a
simpler process of presentation of proof, although enlarged
photographs might serve in the former case.
5. Experimental Graphology.

Graphological exploration has not been conducted solely
by empirical or intuitive methods. Actual experimentation
has been resorted to, although it has only been by slow
degrees that a critical understanding has been evolved as to
the precise problems under consideration and the control
of conditions necessary for satisfactory work. But a grad-
ual refinement of method of experimentation with increas-
ing understanding of the points at issue is, of course, in-
herent in the development of every investigation.

Basal to every attempt at experiment has been the idea
of concomitant variation, change in writing with change in
conditions under which it has been produced. Thus the
concept of graphic variability is found to be essential to the
general hypothesis of grapho-psychodiagnosis and not a
mere embarrassment as many criticshave assumed.
The experiments, as
earliest resumed by Crepieux-
Jamin, us as extremely ambitious in intention
and vague in execution. (9:127f). Yet since the question
at issue is that of the revelation of a personality through

writing it is not surprising that early attempts at testing this
assumption in the experiments conducted by Ferrari, Heri-
court, and Richet consisted in seeking to determine the effect
upon handwriting of suggesting successive personalities to
hynotized subjects. They conclude that "the written ges-
ture is transformed as is the gesture in general" (9:128)
and that, in consequence, proved that variations in writ-
it is

ing are a function of variations in personality. The vague-
ness of the conception of personality renders such experi-
ments of little significance. In a later attempt to render ex-
perimentation more definite comes the device of suggesting
to the hypnotised subject that he assume the personality of
a historic character of very definite individuality. The writ-
ing produced under such suggestion may then be compared
with that proceeding from the character thus simulated !

Crepieux-Jamin recognizing the limitations of the hyp-
notic method, in so far as the subject never completely loses
his personality, rejected the method as unnecessarily com-
plicated and used simple persuasion. The reagent, unac-
quainted with graphology, is first asked to write a given
phrase in his natural way, and then, under definite emotion-
al suggestion, is asked to write it again. The method is
said to be usable only when the subject is both susceptible
to suggestion and intelligent.
It is scarcely necessary to enter upon criticism of the at-
tempt to alter the fundamental individuality of a hand. In
addition to difficulties in the way of manipulating the reagent,

the subjective element in interpretation of results would be
overwhelming. But where deep-seated changes in person-

ality actually occur, as in alternating personalities, a detail-
ed comparison between the handwriting in the two states
should prove most illuminating if there be anything at all
in the graphological contention. But the material at hand
is very meagre. Dr. Prince records instances of "Sally's
inability to write when badly 'squeezed.' She was then
obliged to resort to printing; sometimes both printed and
written characters were illegible. Ordinarily her handwrit-
ing is like that of the primary personalities "(40:561f)
De Fursac, commenting on the modifications from the
normal in the case of mediumistic writing, reports that they
are more apparent than real. They result often in writing
being larger or smaller than the normal writing or in the
slant being modified. But the modifications are for the most
part said to be superficial so that it is not difficult to recog-
nize the fundamental characters of the normal writing of
the medium. Unfortunately, however, we possess only a
few reproductions of mediumistic writing submitted with
copies of the normal chirography. An extensive collection
might prove of great interest.
Experimental graphology must, however, content itself
with a more modest procedure than an attempt to solve all
problems by one device. It has, as a matter of fact, made
slow but real progress three topics which may
t^y attacking
be listed as follows (a^ The limits of objective conditions

as determinants of inmviduality and the specific variations
for which each is responsible;
The range and explana-
tion for variation in the writing of a given individual
apart, of course, from variation due to objective conditions
and the range of variation from one individual to anoth-

er; (c) The range of voluntary control, with specific de-
termination of the graphic elements that may be easily mod-
ified and of those that resist modification.

(a) It is not difficult to list objective factors which might
affect handwriting but it is very difficult to determine the

range of effect. Some of these factors have already been
mentioned. They include such conditions as quality of
ink, quality and size and position of writing surface, fine-
ness or coarseness of pen, objective illumination and tem-
perature, external pressure in the maintenance of speed or
legibility, form of movement employed, finger, wrist, or
forearm. Our estimate of the actual effect of such factors
on writing is, to be sure, somewhat crude. Other factors
more intimately associated with the penman's make-up would
include influence of unconscious imitation (43 ), effect of
written content, effect of bodily fatigue and of specific drugs
such as alcohol. (30 ) Crepieux-Jamin
has reported var-
iations in his writing under changes in weather, fatigue, il-
lumination, visual supervision and the like. (9:136f) He
has also cited changes produced by various emotional con-

(b) This last marks a transition to the second topic,
namely specific variation under specific changes in sub-
jective conditions, variation in mood, emotional excitement,
sensory control and varying degree of impulsion. In such
an investigation, if anywhere, graphologists must find a
justification for assigning specific significance to specific
graphic traits. But variation is two sorts variation from ;

one individual to another and variation in the written prod-
ucts of the same individual ;
there is inter-individual and
intra-individual variation. Can one conclude because the
writing of a given individual varies in a characteristic fash-
ion under given conditions that one is justified in a similar
interpretation of variations in different hands?
The question is a vital one. To give a definite example.
There can be no question that the handwriting of a given
individual varies in size with change in sensory control.
With increased attention to writing we get a decrease in
size, except under certain conditions which need not be
specified here, while with distraction of attention from writ-
ing we get magnification of script. This outcome of exper-
imentation enables us 'to explain some interesting variations

in size of writing for a given individual but what bearing
has it upon variation in size of hand from one individual
to another? The parallelism is not as simple a one as ap-
pears upon the face of it. Is one justified in concluding
that a hand relatively small is a sign of preoccupation with
writing as a process and that large writing is due to autom-
atism of control? The question can be answered only by
an extensive comparison of the handwriting of reagents of
known mental habits. But in any case how set up a group
standard for size, particularly in view of the fact of individ-
ual variation in the expressive threshold ?
If one may parallel group and individual variation, it
would seem that the interpretation of the significance of
size, pressure, slant and alignment should be determined by

study of their variation in the individual and that the sig-
nificance of proportion and continuity should be determined
by inter-individual comparison.
(c) The question of the range of voluntary control for
writing as a whole and for each graphic sign has been test-
ed in the experiments on disguised and retarded handwrit-
ing. A summary of these experiments will appear in the
experimental section. Suffice it to say here that, according
to Meyer, such experiments enable us to determine which
elements are produced under supervision of attention, which
are spontaneous products of motor-impulses. Thus slant,
size, and form are found to be more artificially produced
than proportion, degree of continuity and alignment. More-
over, by noting the specific effect of increased attention upon
writing which is a result of an attempt at disguise we
are enabled to determine just what traits characterize the
controlled hand in contrast with the spontaneous one. Size
and slant, for example, are decreased in the disguised hand,
and there are more breaks in continuity with an emphasis
of the long down-stroke, results which lead us to attribute
to such a hand when it occurs under normal conditions a

higher degree of self-control (inhibition) than we attribute
larger, more inclined, more continuous script.

One word more, it is not easy to say where graphological
experimentation testing the hypothesis that writing may
be used in psychodiagnosis ends and psychological experi-
mentation begins. Certainly the psychology of handwrit-
ing as such should be utilized by a scientific graphology,
while grapho-psychodiagnosis if ever substantiated would
become a part of applied psychology. We
shall find in our
more specific discussions considerable overlapping of fields
of work. One very great difference in point of view should,
however, be pointed out. The psychology of handwriting
is concerned mainly with a study of the writing movement ;

graphology is concerned with the written product. The
former method is highly analytic and has worked out ac-
curate methods for observation as detailed in the Kraepel-
inian studies where precise instruments for registration of
pressure and speed and size are described (13:22:30).
Freeman's writing movement also
fine investigation of the
necessitates a command
of instrumental technique. (16ac).
Psychologists interested in such detailed analyses are apt
to dismiss the graphological program as premature in its
interesteven if not absurdly ambitious in design. Even-
tually,perhaps, the psychology of handwriting may have
something to offer in the way of psychodiagnosis. Mean-
while there is much elementary work
to be done.
For practical purposes, however, judgment must be passed
on the graphic product, not the graphic process. This has
been evidenced by the evolution of handwriting scales as a
pedagogical device. (2:46a). Nor can the utilization of
writing in psychodiagnosis proceed far unless transition is
possible from the movement to the product of movement.
Freeman, however, is reported as directing a handwriting
investigation by means of the kinetoscope which suggests
far-reaching possibilities, one of which may be a convenient
method of studying many individual hands in the process of
making. (Note.)

NOTE See Journal Applied Psychology i, 1917, p. 298.


Pathological Writing.
Graphologists and others interested in handwriting have
j>ng realized that in pathological writing they have a fertile
;eld for work. We
have already seen how Janet urged
of the significance of graphic symptoms
jiat investigation
lould begin with determination of the changes in writing
mt take place under definite pathological conditions. Graph-
ilogists have also realized the value of such material and
sually include in their discussions some reference to patho-
i>gical writing.
Pathologists, approaching the subject from a totally dif-
brent standpoint, have sought to utilize writing in differ-
:itial diagnosis of disease. They have had little interest in
sychodiagnosis as such they have, Trather, been searching

disturbances in the writing of patients.
pr signs of specific
uch a collection as the most interesting one by Dr. Koster
|ad this for its object. He gives characteristic hands for
atients suffering from chorea, hysteria, senile paralysis,
ementia precox, etc. (27) Clinicians who present speci-
icns of this sort in connection with case histories often fail
make any distinction between utilization of graphic ele-
icnts as such and utilization of the written content. Often,
f course, disturbances of attention, of memory, and of
peech- function are evident in the written content quite
part from any specific grapho-motor disturbance. Pen-
ipses are usually analyzed as a product of mental disor-
ers and not scrutinized for evidence of concurrent disturb-
nce of motility.
Not only do workers in this field fail at times to discrim-
nate between graphic and contentual disturbances but, in
eneral, they fail to realize the necessity of presenting the
ormal writing of a patient for comparison with the patho-
ogical. For adequate comparison one should have a series
samples showing the progressive effect of the disease
pon the writing. So inadequate, however, has been the
onception of the requirements for satisfactory comparison
hatmuch of the material that has been published is of very
ttle value.

The same situation is evident in study of pathological
drawings. Nacke (35b) has drawn attention to the need
of samples of normal drawings by the patient for comparison
with pathological productions. He makes the particular
point that inexpertness or lack of training may give a draw-
ing an appearance of being pathological or atavistic in in-
tent although it might be duplicated easily by drawings from
the mentally normal. Nacke's strictures are worth heeding.
None the less, there seems a residue from the work on
pathological drawing that indicates the possibility of utiliz-
ing in some degree drawing in diagnosis. The stereotyped
productions of the catatonics, and the symbolistic pictures
of the dementia precox patient probably have symptomatic
The application of conclusions derived from study of
pathological writing to psychodiagnosis in general is not a
simple one, and certainly not to be settled by a priori con-
siderations. Whether or not pathological writing exhibits
psychomotor correspondences writ beg is a question to be
answered only after elaborate study.
De Fursac, without attempting to pass judgment upon the
outcome of graphological observation, remarks that in any
case the correspondences reported for normal cases do not
hold simply under pathological conditions and he presents
his material in such a way as to make comparison with the
traditional treatment of the graphic signs easy to achieve.
(18 ).
Hirt (23a) makes a threefold distinction of obvious im-
portance but one that is frequently ignored. Quite apart
from physical conditions the writing-act can proceed ade-
quately only if the integrity of the motor apparatus be pre-
served. Hence it is necessary (a) to study the physiolog-
ical conditions of writing and to note those cases of patho-
logical writing that indicate structural changes, gross an-
atomical changes possibly; (b) to work out in detail the
psychophysics of writing, the correlation of determined men-
tal conditions with peculiarities of action; and (c) to con-

sider characteristics of pathological writing that are more
specifically psychological, independent, that is, of physio-
logical conditions.
From the physiological side the investigation of writing
demands consideration of the general conditions of volun-
tary movement and of motor coordination, including the
part played in coordination by visual sensations and sensa-
tions from the moving parts. Clinical experience shows
that insensitive limbs be brought under eye-control.
Skill once acquired only under certain conditions, as
is lost
in ataxia. The writing of the ataxic, both with eyes open
and eyes closed, merits careful study. The psychophysics
of writing involves study of individual variations in both
reflex and voluntary movements. Through observations
of the tendency to and intensity of movements which a man
employs in order to gain a certain end, important conclus-
ions may be drawn relative to his personality. Individual
types of behavior are to be sought in the temporal relations
of movement in the writing reaction-types, where the auth-

or claims to have found experimentally a sensorial and a
motor course; in pressure-types, corresponding to the sen-
sory and motor reaction-types; in rhythmic peculiarities;
and in variations in rapidity of writing and in fluctuations
in rapidity. Numerous problems are raised, as, for ex-
ample, the cause of the increase or decrease of writing-size
when writing is produced with the eyes closed.
In dealing with mental diseases that are characterized
largely by mental symptoms, Hirt appears to find a point of
departure for the characterological study of handwriting.
"How discriminate with security," he asks, "the writing of a
maniac or melancholic from that of a motorly excited or
motorly inhibited man?" In the majority of cases, patho-
logical writing is differentiated from handwriting marked
by personal peculiarities only by the heightening of such
peculiarities. Such comparison of the handwriting of tem-
peramental and insane subjects raises a question which psy-
chiatrists are still debating, the existence, that is, of certain

make-ups which are basal both to character varieties an
anomalies, and to specific forms of insanity which may n
suit in case of strain. (33:42).
In any case it is urged by competent authority, in agre<
ment with the experimental psychologists, that the stud
of pathological writing should not be based on observatic
of the graphic product but that there should be regressic
movements of the p:
to analytic registration of the graphic
tients whoare under investigation. It is thought that: sue
utilization of writing-movements may have actual diagno:
tic significance.

In order to confine our discussion to certain definite issues
letus now consider a few interpretations that have resulted
as a precipitate from graphological analysis. And in order
to give these interpretations background let us attempt a
comparative survey by means of which we may bring into
relationship the different possible methods of approach.
Our procedure will be as follows. Out of the myriad in-
tricacies, the subtle distinctions, given us in the treatises on
the subject we will choose a few graphic elements and sub-
ject them to definite scrutiny from the following points
view: (i) the graphological; (2) the pathological (follow-
ing de Fursac) (3) that of the handwriting expert (fol-

lowing Osborn) and (4) that of experimental investiga-

tion whether motivated by graphological or psychological
terest. We will carry our schematism so far as to attempt
;-;tabular summary of this comparative survey.
The graphic elements chosen for such exploitation are
the following: (I) Size7 or dimension; (II) Pressure and
line-quality; (III) Direction, including slant and align-
ment; (IV) Continuity; (V) Proportion. Some violence
is done the
graphological position by an undue simplification
of it but a certain amount of simplification is necessary in
the interest of a clear-cut presentation.
i. Graphic Dimension.
In this presentation we will limit ourselves largely to

discussion of letter-size. The graphologists tell us that a
"big" hand is a sign of imagination, or ambition, or pride.
The particular form that ambition or pride may assume will
be determined by the general setting in which size is only
one element. Minute writing is a symptom of preoccupa-
tion with minutiae of finesse of miserliness or, sometimes,
; ;

myopia. Again, the general setting is important. Varia-

tions in size are also significant; diminution in size as writ-
ing proceeds indicates ambition or ardor that plays out ; in-
crease indicates waxing ardor.
The determination of whether letters shall be considered
large or small offers considerable difficulty. "Miniscules"
less than two millimeters in height may, however, be cited
as small; and capitals that are less than eight millimeters.
Small run above three millimeters are big and
letters that

capitals that are more than twelve millimeters
high. (Note)
Often, graphologists appear to utilize the capitals alone as
sufficiently indicative of character traits. In this connec-
tion they also make much of the variations in form of the
capital and the possibility it offers for excess decoration.
The characterological interpretation appears to be based
on feeling for size as contributing to prestige; the more

"consequential" a conscious state is felt to be, the more im-
petus toward "large" expression.
2. Let us turn now to pathological writing. Is increase or
decrease of graphic dimensions indicative of any particular
mental condition? De Fursac writes (18:13f) The di-

mension of letters is in large measure a function of the
psychomotor activity or energy. Psychomotor exaltation
or hyperkinesis manifests itself under two different forms
which may be combined in variable proportions, increase in
rapidity of the graphic movements and increase in the ex-
tension and energy of these movements. Specifically, so
far as extent of movement is concerned, we find that aug-
mentation of extent of movement leads to an increase in
thz height of letters. The extent of such magnification is
clearly evident the normal writing of a patient is com-
pared with that produced in a state of maniacal excitement.
Increased rapidity of writing as shown by timed records is
also an outcome of such excitement. The relation of such
increased speed to amplitude is of great interest. When
increase in rapidity does not exceed certain limits it remains

Cf. Graves, S. M. "A Study of Handwriting," Journal of
Educational Psychology, p. 483-494.

compatible with increase in the height of letters, and am-
plitude and speed are associated. But when speed becomes
very great the inverse phenomenon occurs, namely, a dim-
inution in size of letters, a diminution which may result in
certain words being reduced to vague undulating lines,
quite illegible.Often in the same specimen we find both
manifestations of hyperkinesis, increase in height and
diminution of movement due to excessive speed.
The enfeeblement of psychomotor activity manifests itself,
diminution in the height of letters, conjoined
in general, in a
not with increased rapidity but with retardation of the
graphic movement. Specimens are given of such decrease
in sizeunder conditions of melancholy depression. (18:15)
Sometimes a sample of writing from the same patient
shows great variability in size. Fatigue, for example, may
lead to writing that becomes progressively more diminished
in amplitude, while under the influence of automatism writ-

ing increases in size, a fact strikingly evident in stereotypy.
Extreme variability in size may be the outcome of variation
in speed or it may be determined by diminution in the power
of attention. (18:19)
Specific mental disorders furnish examples of such shifts
in size. Thus the writing of the dementia precox patient
may be normal in size or very large or very extenuated de-
pending upon the dominance of automatism or hyperkinesis
or fatigue. (18:151). Even the manic does not always
produce greatly magnified writing; sometimes irregularity
in size is more characteristic than is increase in dimensions.

(18:198). Very great decrease in size from the normal
may occur in the case of melancholic depression, such de-
crease being greater in spontaneous writing than in writing
under dictation, because of the greater mental effort involv-
ed in the former case. But, as before, irregularity in size
testifies to the disturbance of attention.
(18:21 If.)
Other on pathological writing are in pretty fair
agreement with de Fursac. Both Koster (27) and Hirt
(23a:399) reproduce specimens showing magnified writing

when the patient is in his manic period and very minute
writing produced under conditions of depression. Decrease
in through fatigue is substantiated. Koster reports,
also, an increase in size of writing resorted to in an uncon-
scious attempt to mask lack of motor control a device that
I have noted in elderly people.
Gross (22:498), following the more exact technique of
the Kraepelinian investigations, found retardation of speed,
reduction in size, and sub-normal pressure during depres-
sion in circular insanity. Writing characters became pro-
gressively smaller instead of showing the more normal in-
crease in size. In mania, Gross reports that results were
less clear-cut. With rising excitement there was, however,
a tendency to increased speed, increased size and increased
Hirt (23a:397f) reports: The melancholic patient enters
upon the writing act with great slowness and with anxiety.
The stroke of single lines is at times surprisingly weak and
the letters not seldom exceedingly minute. The maniac, on
the contrary, seizes the pen boldly and dashes off the given
proposition in large energetic strokes. On the mental side
the melancholic gives a picture of inhibition, pedantry, anx-
iety, poverty of thought, self-depreciation the maniac ex-

hibits want of consideration, thoughtlessness, incoherence,
3. Letter-size, the expert informs us, is largely dependent
upon the writing system which has been learned. Varia-
tions in this respect are not significant in identification of
writing unless they are extreme. Many external factors in-
fluence letter-size. Thus
the fineness or coarseness of the
pen with which one writing will influence the size of

graphic product. Often, too, the amplitude of the sheet
upon which one writes is a significant factor. Everyone
produces a microscopic hand in addressing a doll's envelope,
and a large one in labelling an express package. It is fairly
easy to alter size voluntarily and within wide limits.

Spacing, Osborn tells us (36a:149) "is mainly changed
by the slant of the upward or connecting stroke," a habit
which is also dependent upon the system of writing which
one has learned. The old round hand and the modern ver-
tical show greater compactness than a Spencerian hand.

4. Size, together with speed and pressure, is a graphic
element that has been subjected to considerable experimental
observation. The Kraepelin studies (13:22) have given
particular attention to it and Freeman (16c) has contributed
a detailed analysis.
There are some interesting relationships observable be-
tween and speed. A graphic rhythm develops in which
there an attempt to keep the time element constant for a

given form even under changed conditions of size. In gen-
eral, we find increased size correlated with increased speed :

there also progressive increase in size as writing contin-

ues, closely related again to developing speed. As attention
is withdrawn from there is an increase in size, par-
automatic writing. Writing that is pro-
ticularly evident in
duced with the eyes closed also shows, normally, an increase
in size, although there are many exceptions to this state-
Decrease in size of graphic movement is an outcome of *

lessened speed or of INTENTIONAL increase in speed.
It occurs, in general, whenever effort is involved in handling
the situation. The direction of attention to the writing
movement as in disguised writing leads to a decrease in
size,although this tendency may be overcome by voluntary
increase of dimensions and in exceptional cases the slow
writing approaches the conventional standard and therefore
becomes larger.
Cutting through these results we find, moreover, small
writing as an outcome of graphic expertness. Illiterate

writing is large because of lack of motor control.
All of the above statements, it should be observed, refer
directly to changes in the extent of graphic movement when
we are dealing with a particular individual under given ex-

perimental conditions. To what degree we are justified
in attempting to apply any of these experimental findings
in an inter-individual comparison of hands is very doubt-
ful. An extremely large and free hand may, however, in-
dicate general freedom of impulse while an abnormally
small hand would lead to suspicion of the presence of in-
hibitory tendencies whichmight vary considerably in nature.
Small writing may be due to excess of control or to economy
of effort as an outcome of practice and skill it may indicate

self-consciousness and inhibition or it may evidence expert-
Dearborn (11, a) in a series of experiments in which a fig-
ure was learned by motor tracery found that concentration
on the conscious movement-sensations led to decreased extent
of movement. He concludes that the conscious movement
sensations are inhibitory in function. There are, he thinks,
two phases of kinesthesia, one unconscious and actuating,
the other conscious and inhibitory in function. From this
it may follow that
large writing is, in general, produced by
the less controlled, more automatic penmen, while small
writing is indicative of concentration on the writing move-
ment or, perhaps, on the external product.

of t- se

ed rected
a n sat
?s ojivS o ra S a fl-C

gSSgS|5-| i||I
is j|o^<u^s.,-3fl
ei J|4|a|l-U j f; *=JI

.52 oJ)-" a> O. W^-?-H

III II 111 1 11 a

2. Force of Movement; Pressure, Line-Quality.
i. The particular graphic quality with which we are
dealing in this section is somewhat difficult to define. The
reference is to intensity or strength of movement as indica-
ted by heaviness or delicacy of the line-quality, its smooth-
ness and regularity. Involuntary placing of emphasis, as
distinguished from conventional shading, is in question.
The general graphological assumption appears to be that

strong firm heavy lines are the outcome of actual pressure
against the surface of the paper and that this, in turn, is an
expression of strong will-impulse. Force of will is deduced
from emphatic and firm movements weakness of will from

delicate and tremulous line-stroke. Transitional forms oc-
cur; persistence is shown by regularity of pressure; spurts
of energy and force by abrupt pressure. The club and stac-
cato strokes as they appear in the crossing of the "t" or in
punctuation marks are thought to be significant.
The explanation for such interpretation is cited as self-
evident; namely, that strength and energy of will express
themselves in forceful and energetic movement. It is, how-
ever, observed that heavy wide lines in contrast to fine trac-
ery are not always the outcome of resolute movement. The
writing materials, such as the consistency and quality of ink
and quality of ink and of paper, the kind of pen used, and the
position in which it is held, obviously condition line-quality.
A very thick stroke points to materialism; while a writing
in which there is no distinction of ground and hair stroke

(the so-called "teigig" or "pateuse" hand) indicates sen-
suality, a love of physical pleasures. The reason
for con-
necting materialism with unusually heavy stroke is not
given it is, probably, purely analogical. Preyer states that

the interpretation of the "teigig" hand is supported by ex-
perience but that the explanation is in doubt. The explana-
tion sometimes suggested is that such a writing is produc-
ed by holding the pen at a very low angle with reference to
the paper and that such holding of the pen is itself indica-
tive of indolence and relaxation.

, From the pathological standpoint we are told that var-
motor energy are reflected in thickness of strokes
iations in
but in very different forms, depending upon pathological
condition. Hyperkinesis produces, in general, "weighted"
writing, the result of which is an increase in thickness of
stroke. The pressure is often so great that the pen pierces
the paper. When, however, speed passes a certain limit,
such pressure is less evident.
Dimunition in energy of movement may have contradic-
tory effects it may result in writing that is abnormally heavy

or in an excessively fine stroke. The former result appears
when both reflex and voluntary reaction are enfeebled as in
epilepsy ;
the latter when the psycho-motor inhibition affects
voluntary contractions only, as in melancholia (18:15,89).
Variability in pressure may result from failure to graduate
pressure in consequence of attention so enfeebled as to fail
to distinguish between the principal and the accessory lines.

Gross, from actual registration of pressure, reported sub-
normal pressure for the graphic movements of patients in
the depressive period of circular insanity (22:498). In
mania there appeared great fluctuations in pressure often ;

extreme pressure was associated with extreme speed and
size as a manifestation of rising excitement. (22:509).
3. Osborn's discussion of line-quality and pressure is en-
lightening. Line-width is dependent to a great extent upon
pen-position. A nearly vertical position gives a fine line of
nearly uniform width without shading and with a tendency
to a broken effect. It is usually associated with finger move-
ment. A
nearly horizontal position gives a broader line,
with frequent shading, and is often associated with free arm
movement. Pen position "can be determined by the loca-
tion of the emphasis of shading." Because of variation in
pen pressure on the nibs of the pen we get, when writing is
viewed under the microscope, three classes of writers:
"those who make the majority of pen strokes rougher on the
left, on the right, or those whose strokes show uniformity

right slant writing more frequently shows an excess of
roughness on the right and lower side of pen strokes"
"The character and extent of the roughness of the line edges
are greatly changed by changes in the character of the sur-
face of the paper, in its sizing, and in the materials of which
it is made. The result is also affected by the character and
condition of the ink used and by the rapidity, direction, and
weight of the stroke." (36:131f.)
The involuntary placing of emphasis is one of the most
personal characteristics of writing and one that "almost
baffles simulation." "The weight of hand, graduation of
pressure, and placing of emphasis radically change the ap-
pearance of a writing as a whole without changing the form
in any way," (36). There may result a hand that sug-
gests strength, one that the record of rapid, nervous move-

ment shown in irregular broken lines or one that shows in

the heavy, ragged, uneven line lack of skill and constant
variation in pressure. Pen pressure reveals the degree to
which writing is free and unconscious or labored and halting.
4. In the experimental investigation of writing, distribu-
tion of pressure in graphic movements is one of the prob-
lems which has been attacked by the Kraepelinian methods.
Gross reports a distinctive curve for every subject tested
but warns the reader that this curve can be detected only by
instrumental analysis. (22:555). Diehl reports that light
pressure and high involuntary speed may coexist. For
example, practice leads both to acceleration of graphic speed
and decrease in pressure. VOLUNTARY increase of
speed accompanied by increased pressure. The relation-

ship between speed and pressure is somewhat indirect; in-
crease in pressure is due to increase of effort of will (Antrieb
or Anregung) ; zeal for work is indicated by rising pressure

Hirt's investigations (23a:370) indicate that writing
pressure obeys certain fundamental physiological and
psychological laws. It increases (i) in a given direction
of movement; (2) under influence of rhythmic tendencies;

(3) at conclusion of a series of
movements (final em-
phasis). It is, however, impossible to determine variations
in pressure from bare observation of writing-product. Only
in part does it parallel thickness or width of line-stroke.
Such divergence between actual pressure and line-quality is

involved in the structure of the pen-point, since strokes
which are perpendicular to the transverse of the nibs of the
pen are necessarily heavier than those which parallel

direction the moving pen point, even though the pressure
be the same in the two cases.
Apart from general laws governing pressure, individual
differences are apparent in the distribution of pressure. Two
main types are observable, correspondent to the motor and
sensory reaction types. The first or motor type makes the
writing movements in one impulse; the second, or sensory
type, fractionate these movements. The impulses of the
motor type are simpler, more continuous, more uncontrolled
than those of the sensory type. The first make movements ;

the second, signs. Writing size, duration, speed, and pres-
sure vary from one part of manuscript to another. At the
beginning, writing is proportionately small, slow, and weak
in pressure. As writing continues there is an increase to a
maximum. Each line is a unit in itself as well as a part of
a bigger whole. Fluctuations in pressure give evidence of
renewed will-impulses. The attempt to produce writing of
fine quality causes more than the usual fluctuation in speed,

size, and pressure, for attention is on the form of the indi-
vidual letter. Writing becomes more uniform in proportion
as allowed to proceed automatically. (23a:383)
it is

Meumann on the ground of difference in degree and dis-
tribution of pressure distinguishes three types of writing
characteristic of men, women, and children respectively.
The concentration of attention upon writing movement
causes an increase in pressure as is evident in disguised and
retarded script, and, in general, increase in effort means in-
creased pressure.

S,fJ of
nils |1'
8*11 * '


S?ils &8 sl

w S
2 .

| || ||g


- -.

^ CO ftraS O 0,

IlII il il

3. Direction. A. Slant.
I. The graphological interpretation of slant is pretty uni-
form. Slant to the right is natural and spontaneous. The
degree of slant indicates impulsiveness, emotional suscepti-
bility the greater the slant the greater the emotivity.
tical writing shows self-control, with the head ruling the
heart. Back-slant is indicative of diffidence , reserve, a
masking of the self, which may be carried so far that it
shows disguise of the self, or even deceitfulness. Excessive
slant to the right is found in the chirography of novelists,
artists, and women. Verticality characterizes the writing
of scientists and thinkers. Actors, diplomats, politicians
may slant their writing to the left, and such slant may also
be indicative of pathological hysterical tendencies or of
criminal tendencies. Great variability in slant is thought
to show variability of mood. Extreme right slant may in-
dicate pathological lack of control.
Three suggestions are found in the literature of the sub-
ject as to the possible explanation of the correlation of slant
with various emotional temperaments.
(a) Preyer's (39) explanation is the common one. Nat-
ural writing slants towards the right as shown by the re-
version to such slant on the part of those taught a vertical
system. Vertical and left-slanted writing demand more
time and effort than natural writing and, therefore, indicate
control and inhibition of natural impulses. Such writing is
evidently self-conscious. Vertical writing may have been
acquired during school years, but is usually replaced by a
more rapid form of writing except in the case of those in-
hibited individuals who refuse to permit themselves to follow
natural impulses and who continue to obey the compulsion
of school or other authority. Back-slant is taught in no
school and utilization of such an uncomfortable method of
writing shows impulse toward concealment or repression,
or it may indicate vanity.
(b) Schneidemiihl (42) has recourse to the general prin-
ciple of expression,
namely, that friendly, objective interests

are manifested by centrifugal outgoing gestures and atti-
tude; and the erect, or withdrawn, posture is expressive of
emotional withdrawal or reserve. In other words, slant to
the right likened to eccentric attitude or gesture vertical
is ;

or and concentric posture.
left slant to erect

(c) Klages (26a) utilizes the principle of spatial Einfiih-
lung in his characterological interpretation of slant. nat- A
ural slant to the right is not considered significant but the
production of a vertical hand is held to be indicative of
stability and self-control inasmuch as it reveals a feeling for
space symbolism which associates fixity and self-mastery
with erect position. Back slant is cited as a sign of extreme
emotionalism with actual repression, rather than as an indi-
cation of extreme coldness of nature as many graphologists
The graphologists cite as evidence of their contention
specimens of writing from persons of known characteristics.
The prettiest bit of evidence is furnished by Preyer who
claims that with advancing age and the loss of emotive sus-
ceptibility, writing formerly slanted shows a tendency to be-
come vertical. His material included two thousand letters
from his father which showed in their sequence an increas-
ing verticality. Furthermore, Preyer cites a case of slant
shifted towards verticality during a period of stress de-
manding self-control and concealment on the part of a young
woman of his acquaintance, a verticality preceded and fol-
lowed by slanted writing.
2. De Fursac tells us that slant is often modified in path-

ological writing, the normal inclination toward the right be-
ing replaced by vertical or back-slanting writing. Some-
times this modification is systematic; the patient seeks to
disguise his hand or give it a touch of originality. At other
times, generally when there is weakening of attention, such
shifts in slant are transitory and casual. In general, slant is
extremely variable in all conditions in which attention is pro-
foundly disturbed. Great variability in slant from right to
vertical and left is cited as characteristic of writing produced

in the post-paroxysms of epilepsy, and in the hyperkinetic
state in manic-depressive insanity.
3. The handwriting expert (Osborn) reports
that slant
is very largely due to the system of writing that is learned

originally. Moreover, such mechanical factors as the gen-
eral position of the body with reference to the writing sur-
face, the position of the paper on the table, and pen-position
are influential. The graphologists recognize, of course,
such factors but consider them accessory rather than basal.
In this connection one may quote from Freeman
(16,d:130) "Irregularities in slant are due to the fact that
in making succeeding strokes the hand or arm is not in the
same position. Sometimes the variations in position and
the accompanying shifts in slant occur frequently and at ir-
regular intervals and sometimes the slant is uniform for a

number of words, or even lines, and then there is a sudden
change. There is also one other type of change in slant
which is due, paradoxically, not to a change in the manner of
holding the hand or arm but to the maintenance of the
same position. This is the increased slant which occurs at
the end of the line."
4. Considerable experimental work has been done on slant
of writing. McAllister estimated from actual experiment
the speed of movement in the different quadrants and found
that movement to the right is more rapid than vertical writ-
ing and that left slant is slow and difficult. "Free full fore-
arm movements in a horizontal plane are made more rapidly
towards the body than away from it, up strokes taking more
time than down strokes." (45:76) Overestimation of
distances arises from increased muscular effort and irregu-
larity of slant may grow out of conflict between eye and
muscle sense.
Experimental graphology has shown that in attempts at
disguise of writing, shift of slant is one of the first points
of attack, the common shift being, of course, from
right slant
to a vertical or back-slant. In general, increase of atten-
tion to writing results in less slant from the vertical. Starch

(43,a) demonstrated that unconscious imitation operates to
change slant when one is writing from a copy.
It would seem as though evidence were pretty complete
that vertical writing, print, and back-hand are slower (Note),
less natural and comfortable than a slant toward the right.

Verticality and back-slant indicate greater motor tension,
greater conscious control, with evidence of inhibition. So
much is plain. The real question is why anyone writes
these hands when he might embrace the greater graphic
comfort of a natural hand. Vertical writing has, of course,
been taught at various times. It is, also, the accepted style
for certain professions. But the natural inclination for
those who have
acquired a vertical hand is to modify it as
soon as pressure is released. Retention of vertical habit
would evidence a conventional, controlled type of person
who follows the prescribed path. Occasionally, vertical

script might be adopted by one who had been otherwise
taught, through an impulse to imitate or because its legibil-
ity makes strong appeal. Whether vertically is ever adopt-
ed because of spatial symbolism is an interesting question
but one not easily answered.
Why, however, does anyone write the awkward and un-
comfortable backslant? This style is taught in no school
and is advocated by no system of writing. It might origi-
nate, of course, through imitation and quite possibly it may
be at times expressive of affectation and self-consciousness.
But that these explanations are not sufficient in all cases is
evident from reports from individuals who write a reversed
hand. Quite often they report that they deliberately adopt-
ed this hand to relieve the strain experienced when writing a
more usual type of script.
Another factor in its production, which has never been
canvassed, is its relation to ambidextral tenden-
cies. Definitely left-handed persons often write very fluent

NOTE The investigation of Graves (loc. cit. p. 490) does not
confirm this statement so far as back-hand is concerned. He found
back-hand most rapid but least staple of all slants.

right-slanted script, but shiftovers who have been compelled
to change from the major to the minor hand, and those with
a tendency to ambidextrality show signs of lack of smooth
coordination in writing or have recourse to a drawing move-
ment which results in print instead of cursive script. Some-
times the ambidexter indulges in reversed script. In a
number of instances I have experimental evidence of ambi-
dextral tendencies in the writers of a back-hand. Further-
more, in connection with measurements on bone lengths of
right and left arm, I found that of fourteen individuals
whose measurements gave little difference in bone-length for
the two arms, five wrote either a completely reversed hand
or gave numerous reversals of slant. Two others, who
wrote a reversed hand, gave only a moderate difference in
arm measurement. I have never found but one individual
writing reversed script who proved by tests to be very
strongly unidextral.
It is difficult to make any connection between such observ-
ations as the preceding, and the teaching of the grapholo-
gists. Yet a connection is possible provided that unidex-
tralityand ambidextrality are correlated with different tem-
peramental types, which correlation if found to be a fact
would find its ultimate explanation in the somewhat differ-
ent functioning of the nervous system in the two instances.

! !!M!
M i
ws nnin (Gra


II <D 1-1 .^J *O
y [^

,2 5 o


Oi o> p 5
a S s

o " -
** . fl -

p] o

fl o


B. Alignment.
I. Variations in alignment are also correlated by the
graphologists with general temperamental traits or with
emotional fluctuations. Rising alignment indicates optim-
ism, self-confidence, or ambition falling alignment, pessim-

ism, depression mediated by outer conditions, or sickness.
Convexity and concavity of line indicate waxing and wan-
ing ardor, that fluctuates as work proceeds. Serpentine
lines evidence suppleness of mind, skill in finesse, falsity ;

serpentine words, quick sensibility, agitation, nervousness.
Modifications of these traditional interpretations appear
in the standard texts. Preyer, for example, fails to find ser-
pentine alignment in the writing of many clever politicians
and diplomatists and, on the other hand, discovers it in the

writing of many persons who are totally devoid of such a
make-up. Nor does Preyer find straight alignment corre-
lated with equability of temperament.
The explanation suggested for the graphological inter-
pretation is that of emotional mimicry. In joyous excite-
ment there is an inclination to raise the arms upward, in
general, to aspire. The sad let the arms fall. But Preyer
observes: "These and also many other mimicry signs have
only a superficial analogy." (39:185) He also calls atten-
tion to the fact that what in writing is upward or
downward alignment is in reality centrifugal or centripetal
movement; only when writing on a vertical surface do we
actually get rising or falling movement. Preyer accepts,
largely on empirical grounds, the traditional assertion con-
cerning rising and falling alignment and instances differ-
ences in alignment in a letter of condolence from that in a
letter of congratulation.
The bar of the "t" furnishes another example of align-
ment, and the interpretation of an up-stroke, a down-stroke,
a serpentine stroke and the like is the same as for direction
of line of writing. In this instance, however, alignment is
complicated with variations in extent and force of stroke.

2. DeFursac, commenting on alignment in pathological

writing, remarks that one might expect that pathological
conditions would have the same graphic expression as the
correspondent normal states, that in the maniac pride and
self-exaltation would produce rising lines and that humility
and discouragement would in the melancholic produce fall-

ing alignment. Unfortunately the case is not so simple.
It is, he asserts, impossible in the present state of our knowl-

edge to determine any constant or necessary relation be-
tween alignment and mental disturbances. (18 ) The
following observations are, however, justified:
1 i ) Undulating lines
are significant from the motor side
of incoordination of movements and from the psychic side
of feebleness of attention.
(2) Falling alignment is seen often (but not always)

conditions of motor weakness, in particular in the post-
paroxysmal exhaustion of epilepsy.
(3) Rising alignment appears in the writing of certain
patients who through lack of initiative fail to give their
paper the desired inclination and permit their hand to move
in an automatic fashion. Rising alignment in such a case
is usually combined with a curved form of the line.

(4) The curved form of the line is associated with the
undulating in certain maladies that are characterized by
automatic reactions, notably in the case of the catatonics.
The forearm remains immobile the hand moves around the

wrist as a center. (18:llf.)
Specifically, in general paralysis lines are often more
less undulating, due both to enfeeblement of attention and
motor incoordination. Falling alignment frequently occurs
in melancholic and depressive forms of nervous disease but
there is no fixed rule. In dementia precox, the direction of
lines varies from a perfect horizontal in some to a rising or

falling alignment in other cases without any possibility

establishing a relationship between the direction of align-
ment and the clinic character of the disease. (18:147)

In manic excitement, also, there is no constant relation-
ship of alignment. Rising alignment occurs but so also does
horizontal and falling alignment (18:197), a statement
which also holds true for melancholic depressive insanity.
3. Osborn finds that alignment is largely the result
pivotage of movement. The writing of the illiterate usually
shows an up-hill tendency. The "arm is so held that the
center of motion is so far to the right that as the hand moves
along it is inevitably raised above the general line of writ-
ing." (36:121) Perfect alignment results when the elbow
is the center of lateral movement and the arm at right angles

to the line of writing. With the wrist as center of motion
there may result lines of writing equal to short arcs of a
circle representing the reach of the hand with the wrist at
rest. Most uneven alignment is due to the fact that the
arm is too far around to the right or the paper too far to
the left. Deviation from alignment in individual letters is
often due to the design of letter acquired when writing was
first learned.

4. From the experimental side there is little to report. It
would appear from Woodworth' experiments (48 ) that
vision functions somewhat in control of alignment and re-
sults obtained from writing when the eyes are blindfolded
confirm his conclusions. Often a loss of alignment is the
only noticeable result with loss of visual control.
Writing disguise affords little material so far as align-
ment is concerned. Alignment is an exceedingly variable
element and one which can be manipulated with ease. It
would, perhaps, seem on general principles that falling
alignment might be cited as evidence of inhibitory tendencies
and Klages in fact lists falling alignment as one characteris-
tic of the inhibited hand. In experiments of my own on re-
tarded writing a decided fall in direction appears as one out-
come of excessive control.
The frequently encountered in the literature
assertion is

of experimental graphology that the contents of an emotion-

al letter or other manuscript influence alignment. Such re-

ports are of little import in the form in which they are given
for there is no narration of the conditions under which such
observations were made. Comparison of alignment in
epistlesof contrasting emotional content emanating from
different penmen is of dubious value unless one have, also,
specimens of the normal writing of each. In this instance
intra-individual variability under prescribed conditions is
the point at issue rather than inter-individual variability.
The problem is a difficult one to attack experimentally since

it is not an easy matter to tap emotion for experimental pur-
have tried the following test. First, I obtained from a
number of subjects specimens of their normal writing on the
blackboard and determined the error in alignment. Then at
short intervals I have had memorized and written sentences
of two or more lines each (i) prophesying a gloomy out-
come of the world war; (2) suggesting encouraging pros-
pects in the war situation; and (3) commenting on certain
amusing aspects of food conservation. Precautions were
taken that the first and second sentences should be written
at the same height and relative position on the board and
that the lines should be approximately of the same length
(one meter). Using the natural writing as the standard of
comparison (in every case there was falling alignment) I
found that out of fifteen items (three averages each for five

subjects) the gloomy content resulted in an increased fall
in alignment twice out of a possible five times the cheerful

content in a decreased fall in direction or even in rising
alignment seven times out of a possible ten. The subjects
were, of course, absolutely unaware of the purpose of the
test. They were adults seriously interested in the war con-
A somewhat similar experiment with students, but less
satisfactorily controlled,gave increased fall with gloomy
content three times out of a possible four and one rise in the

contrasting test out of a possible six. Such results are in-
In Part II, observations are reported on variability in
slant and alignment under normal changes in mood. Anti-
cipating conclusions, I may say here that there was some
evidence of increased slant and unstable alignment under
heightened emotional conditions but that these modifications
were so deeply embedded in general slant and line variabil-
ity as to make practical utilization very uncertain. Results
suggest de Fursac's report with reference to alignment in
pathological writing. Under hyperkinetic and hypokinetic
conditions alignment departs from the horizontal but with
consistency as regards direction.
little Rising, falling and
undulating lines are recorded.

j ft
g| 5- 2 s fe.8

w ztfMiij
o_flo)p, bofiflii St!**fg


Ilia t

4. Continuity.
i. The traditional interpretation of the graphic sign that
we may call continuity, or the degree to which letters are
connected within words, is that connected script is sympto-
matic of a deductive habit of thought, and a broken or dis-
connected hand of an inductive habit of thinking.
Careful reading, however, indicates that the terms induc-
tive and deductive are not to be interpreted in the technical
sense in which they are used in logic. The adjective "de-
ductive" seems to imply a practical realistic type of mind in
contrast to a visionary theoretical type. The " deductive"
thinker does not originate ideas but is thoroughly well able
to assimilate and turn to good account the ideas of others;
he exhibits a high degree of practical judgment and is inter-
ested in application. In general, he may be described as
reasonable, systematic, methodical, and prosaic. The "in-
ductive" thinker produces original ideas which are the out-
come of his intuitions and his lively imagination he is theor-

eticalrather than practical, visionary rather than logical.
His feelings overbalance his judgment.
Preyer (39:138f) gives us a fivefold division of hands on
the basis of degree of connectedness as follows: (a) Pure
"intuitive" hand, every letter detached, breaks sometimes
occurring within letters symptomatic of originality, fertility

of thought, one interrupting another no time taken to follow

any to their logical consequences. Cited as the handwriting
of such men as Chautaubriand, Victor Hugo, Mazzini, Verdi.
(b) The hand that is more intuitive than deductive, sympto-
matic of a mind that is productive of new ideas, reasonable
or not, with greater inclination to follow now one idea now
another than to compare ideas in a logical manner, (c)
Equal number of united and disconnected letters, sympto-
matic of possession of new ideas and capacity to unite them
logically union of idealism and realism enthusiasm for the
; ;

new and appreciation for the old; union of judgment and
imagination balanced intellectual type. Within this group,

however, there appears a subgroup in which the breaks

within words occur in an illogical manner separating the
words into bizarre combinations of letters, a trait which is
cited as significant of impracticality. (d) If the hand is
primarily a continuous one with few breaks, the penman is
thought to possess the gift of synthesis with capacity for
proper appreciation of new ideas. This is the writing of
scientists and statesmen who excel in organization, but are
accessible to new ideas, (e) The completely continuous
hand, every word written without raising of pen, words
bound together by the stroke of the "t" and the like, is
thought to characterize the hand of the assimilative type
who is neither critical, original, nor ingenious.

Schneidemuhl, who cites this interpretation from Preyer,
accepts it only with considerable reserve since his own ob-
servations fail frequently to confirm it. This much, how-
ever, he concedes, that the writer of the "deductive" hand
coordinates and renders coherent the material with which he

Preyer's ground for such characterological interpretation
would appear to be empirical, just as Schneidemiihl's dis-
sent is based on specific observations. I have found in these
authors no psychological ground for their interpretation.
Crepieux-Jamin cites the "hachee" hand as indicative of
intuition but also on occasion as evidencing anguish, or cir-
culatory troubles. The connected hand he finds significant
of natural activity and of culture, or of precipitation and
flight of ideas.
2. Turning now to de Fursac's treatment of graphic con-
tinuity, we find that disconnected or even "hachee" script is
found, on the one hand, when movements are hesitant or
deprived of regularity and harmony, particularly in patients
who are afflicted with trembling, and, on the other hand,
when attention is profoundly disturbed. When associated
with tremor, discontinuity may be the natural outcome of an
attempt at simplification of movement. Script very much
tied together is often produced under conditions of extreme
excitement. Not only the letters of the same word but the

words themselves are joined. This tying together of words
may be the outcome of excessive rapidity. But pathological
writing shows no greater extremes of continuity and dis-
continuity than does normal writing. (18:28)
A more or less connected script instead of broken writing
may occur in conditions of depression and motor enfeeble-
ment, in which case failure to raise the pen is due to lack of
energy sufficient to accomplish this movement. Sometimes
in very great affective melancholia the pen loses contact with
the paper and causes a break even within letters. (18:221 ;
3. For the handwriting expert the degree of continuity is
largely a matter of expertness. Osborn writes "With those

who write clumsily or with difficulty the pen is raised fre-
quently to get a new adjustment with most writers, how-
ever, disconnections are more closely related to design of
letters than with movement, and the habit controlling this
characteristic were acquired when writing was first learned."
4. From the experimental side there is little one can say
about continuity. There is plenty of evidence to show that,
in part at least, a flowing connected hand is the outcome of

graphic expertness. Many breaks in writing may be sig-
nificant of nothing more than graphic unaccustomedness.

Klages shows, further, that breaks in writing may result
from motor inhibition and from excessive attention to writ-
ing. This appears from study of disguised and artificial

writing. One may, voluntarily, introduce breaks into writ-
ing but it is impossible to will extreme continuity. Release
of the motor impulse causes increased continuity while in-;

hibition results in decreased continuity.

any connection whatever between these
It is difficult to see
last observations and the traditional interpretation by graph-
ologists of the significance of graphic continuity. The onlv
possibility of alliance would be found in the determination
of a possible relationship between attention
types and a con-
nected or disconnected hand.

.3 g >, fl

^ a ^ o
"a -S
+3 <D



O P"
-^ rt
S o>
W ^
C fl
03 -d 2 te
- 3,
g3 +j a ffi g,


5. Proportion above and below the base line.
1. The authorities are somewhat at odds in their interpre-
!ition of the significance of relative proportion of strokes
ibove and below the base line. Two traditional views em-
hasize two related but somewhat distinct interpretations. A
mg up-stroke is correlated by the first with predisposi-
on for mental activity and a long down-stroke with a pre-
isposition for physical activity. The second point of
iew states that the long up-stroke signifies idealistic make-
p, impractical and out of touch with reality (the real
orld) and the long down-stroke preoccupation with things

laterial. Balanced proportion denotes power of organiza-
on and administration.

Both Preyer and Schneidemiihl are- sceptical of the tradi-
jonal beliefs. Preyer finds in his collection of specimens
lumerous examples of short upstrokes among penmen fol-
:>wing intellectual pursuits from motives distinctly not ma-
i;rialistic. He has, however, more faith in the deduction
f lack of foresight and circumspection from very short, in-
xnplete, and attached down-strokes. Schneidemiihl from
is study of specimens is inclined, on empirical grounds
nly, to assert a correlation of decreased clown-stroke with
npracticality, lack of foresight, irresolution, and failure in
le execution of details. Normal extension above and be-
\v seems symptomatic of practical sense. He
the base-line
'.utions, however, conservatism in such application and notes
lat there has as yet been no
psychological grounding sug-
ested for such an interpretation.
'Pathological writing affords little information concern-
g the trait in question. One outcome of a hyperkinetic
mdition appears, however, in exaggeration of terminal
ops, of capitals, and in excessive prolongation of move-
ents of adduction.
3. In the general analysis of graphic elements by the ex-
~rt,relative proportion is cited as very simply dependent
x>n the system of writing learned. The Spencerian sys-
m was organized on a scale of fifths the vertical system

in thirds. Prior to the adoption of modern Spencerian,
other proportions existed in relative height of small and loop

4. Some
experimental work has been done on ease of
movement body as a center.
relative to the McAllister
found that "Free, full forearm movements in a horizontal
plane are made more rapidly towards the body than away
from it, up strokes taking more time than down strokes."
In disguised handwriting we find considerable attention
given to relative proportion as of considerable importance in
identification of a hand. While changes in absolute size are
very easily produced voluntarily, certain changes in relative
proportion are maintained only with the greatest difficulty;
increase in the length of the up-strokes, for example. In-
crease in length of the down-strokes, particularly in termi-
nal loops, may, however, be imitated with considerable ease.
Hands vary considerably in amount of difference in ex-
tension of small and lower loop letters. Very extreme in-
equality is usually found in minute writing, an inequality
which Klages (26:37f) considers evidence of the presence '

of strong inhibitory impulses which operate in keeping the
minimum letters small. Very long "long" letters in small

script are interpreted as the outcome of intermittent freeing
of the motor impulse. From Klages' standpoint it is possi-
ble tomake a connection between lack of circumspection and
the short down-stroke.

8 <M "O tr.

3 a> a a 5 c


d -o


The preceding survey impresses one with the difficulties
that must be compassed by a scientific graphology. It sug-
that is over-
gests a program for preliminary exploration
whelming. Any attempt at present to utilize graphic prod-
ucts in psychodiagnosis would seem futile except that in
many instances a variation which in itself has little signifi-
cance acquires such through its association with other signs
which suggest the same motor pattern. It would seem

possible to detect in certain hands signs of inhibition,
or retardation. Slow and interrupted movement, small size,
vertical or back-slant, great inequality in length, and uneven-
ness or heaviness of pressure point to inhibitory tendencies.
Light rapid continuous rhythmic hands, slanted strongly in
the direction of writing, evidence lack of inhibition. It is,

of course, possible to grant so much and yet deny all char-
acterological significance. Graphic habits acquired in youth
may be sufficient to account for the free or inhibited pattern.
But that habit is not all-sufficient as an explanation seems
evident from the following. Pathological writing under
prescribed conditions exhibits in the hyperkinetic hand an
exaggeration of the explosive or free hand. Moreover, the
changes introduced into the writing of a given individual
under conditions of increased control or mental effort
encourage the interpretation of certain signs as symptomat-
ic of control and effort in general. In Part II, we shall re-
port a number of experiments that bear upon an attempt to
interpret the significance of explosive and inhibited hands.
Part II


In our previous discussions we have had frequent occasion
to refer to the significance of handwriting disguise. Let us
now envisage the subject at closer quarters.
The determination of the extent to which handwriting
may be disguised is a problem of considerable importance
from at least two points of view. On the one hand, as a
practical problem, of great interest from the legal

point, it arises in connection with the imitation of the writ-

ing of others in forgeries that are not traceries and in the
"masked" writing of the anonymous letter. On the other
hand, from the psychological or theoretical side, the range
and method of handwriting disguise is, as we have seen, of
significance in connection with the utilization of handwriting
n psychodiagnosis.
Our concern is, of course, with the second of these two
considerations. In passing, however, it should be observed
hat the handwriting experts have much to say concerning
he difficulties involved in the identification of handwriting
and the determination of the original of a disguised hand.
They insist upon the need of cautious procedure; they list
he sources of possible error and they warn against the ac-

ceptance of the unsupported opinion of the incompetent and
untrained witness. Mr. Osborn writes, (36;c) "There are
wo main questions that confront the examiner of an alleged
forgery. The first of these is how much and to what extent
nay a genuine writing diverge from a certain type, and the
second is how and to what extent will a more or less skill-
forgery be likely to succeed and be likely to fail in em-
Bodying the characteristics of a genuine writing." These two
questions (i) of the limits of variation in a natural hand
and (2) of the graphic characteristics that may or may not

be easily assumed are of first importance from the theoreti-
cal side also.
significant item of difference between the emphasis of
the handwriting expert and that of the psychologist should,
however, be noted. The expert approaches the problem
largely from the standpoint of the degree of credibility of
the witness testifying in court concerning the genuineness of
handwriting. The
psychologist would press the matter fur-
ther back and determine,
if possible, the reason for the great

individual differences that exist, apart from training, with
respect to observation of handwriting individuality. Furth-
ermore, he is most curious concerning the varying capacity
for disguise exhibited by different penmen and the mental
temperament that lies back of virtuosity in the assumption
of different handwriting individualities. Lastly, he would
ask what the psycho-physical factors are that determine the
ease or difficulty with which different graphic elements may
be voluntarily altered.
The problem of control in handwriting, which we have
emphasized as a basal one so far as psychodiagnosis is con-
cerned, centers about two problems both of which are open
to experimentation ( i )
: The extent to which disguise of
one's habitual handwriting is possible and (2) the extent to
which voluntary control is maintained in conventional writ-
ing as evidenced by the changes that take place in automatic
writing or writing under distraction. In everyday life an
obvious indication of this latter change is the difference be-
tween writing furbished up for state occasions and writing
designed for domestic purposes, in negligee so to speak.
In ordinary writing, control becomes progressively less rigid
as one becomes interested in the content of what he is writ-
ing or as speed of writing increases. The first half of each
word, the first half of a written line, and of a manuscript
give evidence of greater control than does the second half.
The significance of this variation in conscious control, so
often emphasized by graphologists, need not detain us here.
Instead let us turn to the problem of voluntary disguise of

For scientific purposes one strikes the problem at close
quarters by an experimental treatment such, for instance,
as that of Dr. George Meyer. (34a) Meyer approached
the question from four different angles ( I ) Which graphic

characters can be repressed voluntarily? (2) Which can
be assumed voluntarily? (3) What is the result of a delib-
erate attempt to disguise handwriting? (4) How far is

imitation of another hand possible ?
To obtain an answer to question I, a large number of sub-
jects were asked to write as calligraphically as possible, in
true copy-book style of the school-room. Normals were
also obtained for comparison. To obtain an answer to
question 2, definite variations in particular graphic charac-
ters were asked for from twenty-five different reagents.
Question 3 was answered by asking subjects to disguise their
writing question 4 by asking for imitation of specific hands.

From his study of themethods of intentional disguise em-
ployed by unsophisticated subjects, Meyer was able to draw
some interesting conclusions concerning the graphic elements
that are least subject to control, which in the main are pre-
cisely those to which the average observer pays least atten-
I have notes on an
experiment of my own similar in pur-
pose to that of Meyer but developed in a somewhat different
I asked twenty- four
unsophisticated subjects to write a
given verse on an unlined sheet of standard size and quality
in their usual manner. I then requested each of them to re-
write this verse on a second similar sheet but
their handwriting as far as possible. No instructions were
given as to method of disguise. Each subject could take
all the time and
pains that he cared to in the disguise, which
was prepared away from the laboratory. In
selecting my
subjects I choose twelve of each sex. With reference to
age they fell into two groups also, twelve under
years of age and twelve over thirty. The younger group
was, with one exception, composed of college students; the

older group, with three exceptions, of University instructors,
Four of the latter were psychologists. Such a selection of
subjects was dictated by a desire to see whether age and sex
were factors governing success in disguise. The degree to
which a given disguise was held to be successful was deter-
mined by the submission of the series of disguised and un-
disguised writings to sixteen reagents for matching and the
counting of the number of times a disguised specimen was
correctly matched with the undisguised specimen written by
the same penman.
The material obtained in this manner was worked over
with the following questions in mind: (i) What methods
of disguise were utilized by the group of subjects? (2) To
what extent were the individual attempts at disguise effec-
tive as determined by the percentage of failures on the part
of the judges in identification of the disguised hand? Were
the younger penmen more successful than the older ones in
disguise? Was there any difference in the percentage of
successes of men and women?
In an attempt to answer the first question, out of almost
numberless observations that might be made relative to
changes in the graphic characters, tabulation was limited
to the obvious shifts in size, slant, pressure or line-quality,
form, continuity, alignment, connecting-stroke, relative pro-
portion, and i-dot. See Table I.

O rt

S '
7 rt

o <u

5 ^
55 S


A word of comment upon each of these chosen elements is
desirable. A change in size of writing is a frequent out-

come of disguise, a decrease being more common than an in-

crease in size of letters. There in fact, in the given speci-

mens no case of increased size comparable to the extremes of
decrease. The decrease in size of letters is usually accom-
panied by greater compactness in texture leading to a com-
pression in the horizontal extension. This same compres-
sion appears also in a few cases in which the writing is in-
creased in size but, usually, increased amplitude is accom-
panied by a looser texture.
shift in slant was also noticeable in the disguised hand,

usually in the direction of the vertical or backhand. Such
a change is one that readily appeals to the unsophisticated,
although handwriting individuality is but little dependent
upon slant of writing.
The degree to which pressure varied in the natural and
the disguised hand cannot be told with any degree of accur-
acy from the written product. Experiments on pressure
demand, as we have seen, actual instrumental registration.
Certain changes in line-quality were, however, very evident
in a largenumber of cases. In a majority of specimens this
change is in the direction of a heavier line. I do not find,
unfortunately, a record of how many of my subjects used a
different style of pen in attempting to disguise their hands,
but in any case it is unlikely that such a shift accounts for
the uniformity in direction of change.
In consideration of variations in letter-form, the writing-
specimen was scrutinized to determine whether on the whole
there was simplification or conventionalizing of the natural
hand or whether the reagent attempted to disguise his hand
by the employment of superfluous ornamentation or fantas-
tic forms. Recourse to a conventional vertical hand or to
print is one of the most effective means of disguise but it is
more difficult to achieve than a hand decked out with all
manner of superfluous curls. It demands more consistent
motor control. The tabulation given overlooks the many

details of form that would be so carefully noted by the ex-

pert in attempting to prove or disprove the genuineness of
a given writing. Individual mannerisms, tricks of style, are
often revealed in the form of individual letters and one of
the most interesting questions involved in disguised hand-
writing is the extent to which a penman is aware of his in-

dividual peculiarities and the consistency with which he is
able to avoid tell-tale mannerisms. Such observation does
not, however, lend itself to tabulation. Changes in capitals
are more
easily achieved than changes in small letters they ;

are made with a higher degree of consciousness.
A change in alignment occurs frequently but without much
uniformity as to the direction.
The degree of continuity in a given hand is one of its
most distinctive marks. This character is held to be very
largely dependent upon the general smoothness and regu-
larity of the motor impulse, a matter, to a considerable de-
gree, of the original constitution. A break in continuity is
much more easily initiated than is increased continuity, as a
simple outcome of intentional inhibition. It is a much more
difficult matter to release deliberately the motor impulse and
so increase the degree of continuity.
Changes in the form of the connecting stroke occur fre-
quently, more commonly from an angular to a rounded con-
nection than the reverse.
While the absolute size of writing is easily shifted, rela-
tive proportion of parts is pretty constant. The latter de-
pends upon periodicity of effort which is rooted in constitu-
tional rhythm. There are, however, a very great number of
possible observations relative to proportion, among them
the following: Relative proportion of strokes above and
below the relative size of one-space and three-space
horizontal and vertical space relations of the
letters, relative
one-space letters, relative proportion of capital and strokes
above the line. It is much more difficult to vary some of
these proportions than others. For example, from my re-
sults, it appears that a change in the relative size of the one-

space letters is not infrequent, while changes in the relative
proportions of up and down strokes is less often observed.
An increase in the relative length of an up-stroke is said to
be particularly difficult to achieve and my specimens show
only two cases in which such a change was evident. An in-
crease in the relative length of a down-stroke is much more
common, and, in general, an increase in difference in length
is twice as frequent as a decrease in difference in length.

The mannerisms exhibited in dotting the "i" are very con-
stant. This i-dot may be observed from three points of
view; its localization, that is, the distance it is placed above

the line and its position directly above or to the right or the
left of the "i" ; secondly, its form, which varies in an ex-
traordinary number of ways from comma-shaped to wedge-
shaped, not to mention its size and, thirdly, the time of its

making, immediately after the letter itself or after the word
or line has been written. One would need to watch the
penman while writing in order to establish this latter habit.
In the disguises I collected, there are no obvious changes of
localization in the placing of the dot, although in several
specimens there is great variability in the natural hand it-
self. So far as form was concerned there were several de-
liberate attempts to vary it. Bizarre substitutions were
I am
adopted, such as the circle, or a v-shaped figure.
inclined to think that two or three of these changes were
motivated by a knowledge on the part of the penman of the
fact that the dot of an "i" is most characteristic.
Some of the changes just mentioned are deliberate, a
revelation of what the subject believes to be characteristic
of handwriting. The more constant a mark, the less con-
scious awareness of it. Of these deliberately sought shifts,
some are easily manipulated, slant, for example, and change
in absolute size. Others are handled with greater difficulty
because of their dependence upon psycho-physical factors,
as, for instance, degree of connection. Still other changes

are dependent upon the general instruction to disguise the
hand and are not directly willed by the subject nor even

noticed by him. Such are the three general symptoms of
disguise: (i) Instability, (2) Signs of Tension, and (3)
Inner Contradiction.
Let us be more specific. Absolute size is easily changed
voluntarily. But not all changes in size are to be attributed
to direct volition. Increased attention to writing results in
decrease in the size of writing and in increased pressure.
The uniformity with which changes occurred in these direc-
tions is, then, in part at least, an outcome of effort of atten-

tion and not wholly a product of intention. Increased size,
on the other hand, may result from discontinuity of the
motor impulse so that each letter is written as a separate
unit rather than as part of a word. One would expect to
find this increase in size in disguises in which attention is
concentrated upon variation in the form of individual letters.
Frequently, breaks in connection between letters would also
result from such a break-up of the motor impulse.

Alignment and the shape of the i-dot may be deliberately
varied if one chance to know his own mannerisms and if
he can hold his attention consistently to the detail in ques-
tion. Details of form are very hard to change, particularly
in the middle and at the end of a word. The style of a
capital is not hard to shift.
The results of this canvass of the methods utilized in dis-
guise of hands agree very closely with what has been re-
ported by Meyer as the outcome of his investigation and
with the scheme adopted by the Berlin police in their in-
dexing of handwriting specimens as part of their system of
identification of criminals. In this latter system the charac-
teristics of handwriting are arranged in a descending scale

beginning with the elements that are most easily altered and
ending with those that are least subject to change.
The order is as follows (26a) :

Emphasis of the under stroke
Ataxia (unformed and trembling writing)

Increase in proportional length
Emphasis of the upper stroke
Form of the connecting stroke
Decrease in proportional length
Change in single forms
A further point of interest is a comparison of these shifts
that accompany an effort to disguise the hand, with con-
centration of attention upon the act of writing, with those
that are the outcome of distraction of attention from writ-
ing and, in some instances, of completely automatic writing.
The shift in size that is significant of automatic writing has
been somewhat thoroughly discussed in another connection
(14a). Increase in size is a general outcome of increased
automatism, just as decrease in size is an effect of concen-
tration of attention upon writing, unless the latter result in
a complete dissociation of letters and a distinct motor impulse
for each. Adecrease in pressure is also an outcome of
automatic writing but less evidently so than the increase in
size. Completely automatic writing results apparently in
script thatis more continuous than the usual
writing but in
case of incomplete distraction there would be alternate fixa-
tion and release of attention with, probably, increased dis-
continuity. Changes in slant do not occur in automatic
writing as they do in disguised hands, although there seems
to be in some cases a tendency to greater verticality. Changes
in form are in the direction of disorganized or child-like
Between the two extremes of voluntarily disguised writ-
ing and writing produced without conscious supervision lies

the ordinary writing with which graphology deals. It is
evident where one should look for lapse of control. Periodic
fluctuation of attention enables us to anticipate the fall of
the mask at various points. In ordinary writing there is
heightened consciousness and hence greater control (i) at
the beginning of the activity, inscribing, for example, the
first page of a manuscript, the first word of a line, etc. 5(2)

after interruption of the writing activity by paragraphing or
punctuation marks; (3) at variation in the form of the ac-
tivity such as the production of capitals. Controlled writ-
ing is smaller, more vertical, and more regular than uncon-
trolled writing, that is, the same signs appear as in dis-

guised writing but in less pronounced form. Conventional re-
straint becomes progressively difficult as speed of writing
increases. With deepening interest in content, writing be-
comes freer and bolder. Every prolonged piece of writing
shows the shift from conscious to involuntary control, and
in this fact the graphologist finds an opportunity for ob-
servation of certain characteristics of the motor impulse.
Let us turn now to the second question, the success of a
disguise as determined by the failure of the judges in pene-
tration of the disguise.
But before entering upon the question of the success of
the individual penman, a word concerning the varying skill
of the sixteen judges. The range in success runs from only
one correct identification of the twenty-four specimens of
disguised hands to an accurate pairing of eleven specimens
(a record made by a bank cashier). The average number
of correct identifications (and the median
record) is six, or
twenty-five per cent.
About half of the judges were taken from the college
community which produced the disguised hands, and, in
some cases, they recognized a number of the natural hands.
This familiarity with the natural hand increased
slightly the
number of correct identifications. There exists, however, a
very great individual difference in the ease with which
handwriting is recognized even when undisguised and in the

facility with which handwriting specimens by the same pen-
man may be paired. For ten of my judges in this test I
have record of their success in the matching of undisguised
hands. The group is too small to be of much value but the
results of the two tests give a positive 'coefficient of correla-
tion of .41 (P.E., .18).
The outcome of this aspect of the experiment justifies the
distrust on the part of the most careful handwriting experts
of the opinion of the ordinary observer as to the genuine-
ness of a given hand. The chance of error is so great that
the judgment of the amateur can have little weight, al-
though, obviously, the opinion of one may be worth more
than that of another, a matter which could be determined
only by a controlled test. Certainly the confidence with
which a witness or a reagent in the psychological experi-
ment expresses his opinion bears little relation to his value
as an observer and might be most misleading in a trial in
Such strictures against handwriting-identification on the
part of the amateur only 'serve to point the value of the
work of the expert, with his instruments of precision, his
microscope, his enlarged photographs, his multiplication of
observations, and his knowledge of where to look for sig-
nificant variations.
Three of the judges who took part in the test on disguised
handwriting were given a second trial at matching after
an explanation had been given them concerning the signifi-
cant features of writing individuality. They were advised
to ignore changes in size, slant, and form of capitals. Their
increased success was as follows: (A) from nine to twelve
correct identifications; (B) from six to nine; (C) from five
to eight.
So far as the individual disguises are concerned, some
were much more effective than others. Three hands could
scarcely be called disguised since they were correctly paired
by almost every judge. Of the other twenty-one, four were
so well disguised as to wholly elude capture. Eight were
identified by only one or two judges each.

The three who fail completely at disguise write very in-

dividual hands. Their failure was evident to themselves
and they made subsequent attempts to mask their writing
without much
greater success. Of the four completely suc-
cessful disguises, one is a semi-print style; another is a
most clever imitation of a friend's hand included among the
specimens, with which it is matched by four different
judges. Two others show very great shifts in slant and size,
changes which however easily manipulated seem quite ef-
fective in deceiving the ordinary observer. The more con-
ventional and immature hands that approximate a given
system cause considerable difficulty in the test.
Calculating the percentage of actual identifications, on
the basis of the possible number, for the groups of older
and younger subjects respectively (twelve each), we find
it34 per cent, for the older group and 18.3 per cent, for the
younger. The three reagents who completely failed to dis-
guise their hands all belong to the older groups. Dropping
these out and recalculating on the basis of actual to possi-
ble identifications we find the percentage of successful iden-
tification for the older group is 17.4 and for the younger
18.3 per cent.
Calculating the percentage in the same way, but with a
divisionon the sex basis (twelve each), we find the percent-
age of correct identification is 22.5 in the case of the
women and 30 per cent for the men. Again dropping out
the three subjects who failed so signally at disguise (two
women and one man) and recalculating, the percentages run
10.3 per cent for women and 24.7 per cent for the men.
Our numbers are too small and too greatly affected by indi-
vidual records to be of great value but so far as they go
they indicate that women are more successful in disguise
than men, and the younger penmen more successful than the
older. All of the four subjects whose disguises were not
penetrated were women (one from the older group, three
from the younger). Of the eight specimens recognized by
only one of the judges, four were written by men (three of

the older, one of the younger group) and four by women
(three of the younger group, one of the older). The best
records, so far as disguises are concerned, are made by the
young women.
The success of the younger group, particularly those
writing an immature hand, is not necessarily due to the as-
sumption of another graphic individuality. A
return to the
conventional system would cause a confusion of such a
specimen with others similarly motivated. In a personal
letter in which he comments upon the specimens used in the

present test, Mr. Osborn writes me "Writing by those who

have not long been doing writing outside of school is bound
to be simliar in many ways and when such writing is dis-

guised its individual features may be modified and its gen-
eral features remain, which would tend to connect specimens
written by different writers." Yet this is not the whole
story. In a few disguised specimens there is, very evidently,
the assumption of a distinct, yet different individuality. The
most interesting disguises are those in which there occur
such curious changes in style. Some of these disguises
come from the older group and lead to the conclusion that
an effective disguise is much more a matter of the individual
constitution than of age or even sex.
It has been held that ability to shift handwriting individ-

uality is akin to ability in acting. But we have as yet no
analysis of what traits characterize the dramatic type, al-
though Holt suggsts (25:35) that "The actor's is merely
the excessively mercurial and labile character." From my
knowledge of my subjects I should say that those showing
much facility in the adoption of another chirographic indi-
viduality, were, in the main, much more adaptable, more
pHable, than the others. There is, however, one rather
striking exception to this statement. This reagent
a girl
of the younger group is very visual in type and talented
in drawing and fine handicraft. She took pleasure in pro-
ducing for me an amazing variety of hands. Personally,
she is of a distinct and somewhat inflexible individuality

who yields slowly to social pressure. She is artistic, rather
than imaginative.
Four of the subjects in this test were also reagents in my
experiments on control processes in handwriting (R, B, S
and D) (14a). For these I have a fairly complete analysis
of their general procedure in writing. Of the four, two (R
and B) were highly successful in their disguises and two
(S and D) were inapt. D was particularly poor and that
in spite of the fact that she was probably more aware than
any other person who attempted the disguise, of the tell-tale

points in chirography.
It is certainly significant that the alignment of these sub-

jects in the test on handwriting disguise tallies with that
found in the earlier experiment. R and B belong to the so-
called "motor" group; D and S to the "sensory." Charac-
teristic of the first group was the high degree to which
writing was turned over to automatic control; characteris-
tic of the second was the maintenance of conscious writing

control, usually accompanied by a vivid sense of kinesthetic
sensation. For the latter there is consciousness of muscu-
lar effort in writing and evidence of motor inhibition. For
the firsttwo the act of writing is successfully organized and
the motor impulse smooth and effective. (14a:14o f.)
Interpreted on a conventional habit-basis one might per-
haps expect the first two subjects to be less expert than the
other two in disguise of the hand. But undoubtedly our con-
ventional views of habit need reconstruction, especially along
the line of ease in habit-breaking and the relation of this to
the mental organization and constitution as a whole. Very
possibly the cue to the interpretation must be sought in the
smoothness, effectiveness, and lack of conflict in the motor
impulses themselves, which would facilitate both habit-for-
mation and quick readjustments.
A dramatic reaction to the instruction to disguise one's
hand, in which one initiates and then yields confidently to a
graphic-motor pattern somewhat different from his habitual
one, is more effective in disguise than is an effortful dis-

integration of graphic details, with a constant effort at inhi-
bition of habit. That both kinds of disguise may be suc-
cessfully achieved, is, however, evident. Psychologically
and practically they are of differing interest. The effortful
disguise, although it may conceal its source effectively, will
give evidence of not being a natural hand by inconsistencies,
by retouching, and by the presence of fantastic forms. This
type of disguise is, possibly, that most often found in the
anonymous letter. The dramatic disguise will be less evi-
dently a disguise and in its most successful forms points to
an interesting mental type. It occurs in certain forms of
Meanwhile we note that Klages cites versatility in the
shift ofhands as characteristic of the fluidic personality.
Indeterminate personalities have less to control or conquer.
Graphic virtuosity is evidence of histrionic ability or of the

splitpersonality of the hysteric. The subject is worth inves-
tigation both in connection with a study of one type of crim-
inal the forger and investigation of the hysteric temper-
ament and of the double personality. Several possibilities
of application suggest themselves in connection with the
utilization of writing in diagnostic tests.

In the preceding discussions we have had frequent oc-
casion to emphasize individual variability in writing. The
following study seeks to investigate the range of such varia-
tion in the writing of two subjects (I and II) and, in par-
ticular, to correlate variations in alignment, slant, and size
with changes in emotional conditions.
The material was gathered in the following manner For :

four months (October-February, 1913-14), in connection
with another piece of experimental work, my collaborator
and myself kept a daily handwriting record under standard
conditions. At approximately the same hour of the day,
usually at the same laboratory desk and with a pen of stand-
ard number, on a sheet of a given size and quality we (i)
recorded the name, date, and pulse-rate; (2) wrote a para-
graph on weather conditions; (3) gave a description of our
physical condition; (4) recorded in detail our mood; and
(5) summarized our interests for the day, rating the strength
of each on an arbitrary numerical scale.
Our general problem was the gathering of material for a
study of fluctuation in interests and in general patterns of
consciousness. But I wished also to secure a standard series
of writing specimens for analysis of graphic changes corre-
lated with variability in condition. At the time, however, I
hc.d in mind no with reference to the par-
definite questions
ticular points to be studied in connection with emotional
variability, a condition which enhances the value of the
material for the present purpose.
The series afford excellent stuff for such a study as I
now wish to make, for the mood records of both subjects
show considerable variation. There are, however, different
factors concerned in the two.

For Subject I, the period under study proved to be one of
very great emotional stress and strain, highly depressing in
nature. On two occasions there occurred objective shocks
of considerable intensity. Subjective fluctuations were also
recorded. Periods of great absorption and interest in work
were described and some of restless excitement but gay or
happy moods were rather infrequent and in intensity were
not comparable to the contrasting moods. The physical
condition was poor, inducing rapid pulse, nervous irrita-
bility, and motor incoordination.
For Subject II, the mood changes were largely condi-
tioned by varying physical conditions. The usual condition
was one of energy and high interest in work, with occasional
fatigue and lapse of interest in things in general. Three
periods of low physical vitality were recorded, one incident
to an attack of la grippe, the second an outcome of vaccina-

tion, the third the effect of an accident. A characteristic
report of Subject II was the more or less periodic appear-
ance of a day-dreaming mood, subjective in tone, pleasant,
and accompanied by relaxation. A vacation period of ten
days was toned by a sentimental and highly pleasurable
My general procedure in checking over records was as
follows Under the appropriate date I listed the description

of the physical condition and of the mood; I also listed the
pulse and energy record, although I found these less com-
plete than I would wish. I measured in millimeters and en-
tered in my tabular summary the horizontal extension of
each autograph, the height of three capitals in the name,
the height or length of one three-space letter, and the slant
of these same four letters. There are ninety autographs
available for Subject I and one hundred and two for Subject
II. The specific ways in which this tabulation was utilized
will become apparent as the report develops.
Alignment was estimated in the following way: I meas-
ured with care the departure from the horizontal for the
particular line on each sheet which recorded the mood,

choosing this line because if content influences alignment,
as graphologists report, it should most typically represent
the condition for the day. This line was written about mid-
way of the sheet another reason for choosing it as repre-
sentative. All the record sheets were of the same size
(21.5 x 13.9 cm.) but on account of variability in marginal
spacing there are some slight differences in line lengths.
After measuring the departure from the horizontal of the
line in question a general estimate of the tendency through-
out the whole record was obtained. The measurements
throughout are somewhat crude but probably sufficiently ac-
curate for the purpose in hand. A
more accurate measure-
ment of writing amplitude by means of the curvimesser is in
According to graphological tradition, alignment is deter-
mined to a large extent either by general temperamental tone
or by the mood dominant at the moment of writing. The
straight line characterizes the person of equable disposition ;
the up-tendency appears in the writing of enterprising, hope-
ful, optimistic penmen; and down-alignment in the produc-
tions of those of generally pessimistic inclination or of tem-
porary depression, physical or mental. Cases are cited in
the literature inwhich great variations in alignment result
from mood. Slant is also cited as dependent upon
shift in
emotional instability. Size, so far as related to mood,
would show increased amplitude in energetic, hopeful states
and decrease in size in depressive states.
To determine whether the mental condition had any effect
upon these three graphic elements, I checked over my tabu-
lation of moods and selected, wholly at random, six for each
subject, under the three following descriptions:
I. No
emotional toning to consciousness. Neutral.
II. Consciousness toned with gayety, happiness.
III. Consciousness toned with depression, melancholy, or

physical sickness.
Only five entries under the second rubric were discovera-
ble for Subject I.

O & o.S

o o vo *c o *


-f + 4- > Q
i H


15.-T- 1

Tables II and III summarize the results. Size and slant
measurements were made on the autograph as explained
above. Under extension, the horizontal extension of the
whole name is entered; under height the SUM
in millimet-
ers of the measurements on the four chosen letters; under
slant the average degree of slant from the vertical for the
chosen letters.
Someinteresting facts emerge from study of the individ-
ual records. First, that of great variability in all the ele-
ments measured. Thus for Subject I, length of name
ranges from forty- four to eighty-three millimeters extension ;

of the first initial from ten to twenty-nine millimeters aver- ;

age slant on the chosen letters from sixteen to forty-two
degrees. For Subject II, the length of name ranges from
fifty-three to one hundred and ten millimeters, extension of
the initial capital from five to twenty-six millimeters; and
average slant from less than five to nearly fifty-three de-

grees. Both penmen show, I suspect, an unusually exten-
sive range of variation for the traits measured. Both show
on occasion a tendency toward excessive slant; toward the
left for I toward the right for II.

The tabulation of the specific records indicates that results
from Subject II ofler some confirmation of graphological
principles. Average extension, height and slant are all in-
creased in a pleasurable state whether comparison is insti-
tuted with the products of the neutral state or with those
of depressed conditions. Not only is this true, but, further-
more, depressive states also show, in conjunction with de-
crease in amplitude, an increase in slant as compared with
the neutral, quite in accordance with graphological expecta-
tion. Alignment presents less straightforward results. In
general, Subject II produces a very irregular alignment.
One can assert this much only, that there is a trifle greater
tendency to up-alignment when consciousness is pleasurably
toned than is the case otherwise.
Results from Subject I are very different. In this in-
stance the records produced in the neutral or objective-

minded states exceed the others in extension and slant ;
is very little difference in result between depressive and gay
states. Alignment which is normally rising shows a con-
siderable inclination to fall in depression.
There was undoubtedly some difference in the significance
of the so-called neutral state for the two subjects. For I,
the real contrast between the first and the other two condi-
tions is expressed by the terms Objective- Subjective (possi-
bly extroverted-introverted). The objective state of mind
is characterized by intense interest and absorption in work;

it isunemotional but probably pleasant and is more charac-
terized by energy than the contrasting states are. For Sub-
ject II, the opposition is between indifferent and affectively
toned states of mind. States of great interest in work and
of physical energy are happy states. Depression is, usually,
the outcome of sickness. For both subjects, however, a re-
lationship exists between energy and the resulting graphic
expression. With high energy there is, in general, an in-
crease in scope of movement.
Anticipating a distinction to be emphasized later, we may
say that Subject II writes a typically explosive hand,
large, light, rapid, centrifugal, tied together, with excess
of occasional movements and little distinction in proportions.
Subject I writes an inhibited hand, small, somewhat slow,
centripetal and broken, with great distinction in relative pro-
portions. Alignment is the contradictory symptom in each
of these hands, since I shows a tendency to rising align-
ment and II a tendency to falling alignment. Release of
tension such as occurs in states of unself-conscious absorp-
tion in work shows in I's case in the production of a hand
more explosive than usual, increased in amplitude, speed,
and continuity. Attention is diverted from the graphic
product. II, naturally of a more objective mental set, ex-
hibits under pleasurable excitement increasing impulse but
under depression, restraint of movement.
While such results are of very great interest they sug-
gest from the practical side the difficulty inherent in any at-

tempt at utilizing size and slant apart from an understand-
ing of the particular case involved; only under prescribed
conditions can increase and decrease of size and slant be
revelatory. Fluctuations in alignment offer still greater
difficulties in interpretation.
As a second method of testing the records, I selected ma-
terial for tabulation, starting not from the mood-side but
from the measurements I had listed under extension. I took
the highestand lowest ten per cent, of these measurements.
Table IV. gives the tabulation with correlative data.
This tabulation reveals little more than the earlier ones.
As before, a greater agreement with graphological principles
is manifest in the record of Subject II than in that of Sub-

ject I. But the records of both subjects show numerous ex-
ceptions to the general point of view. When energy is high
and there is an active, working, happy mood, there is a ten-
dency for Subject II to produce magnified writing but the
same tendency is evident in nervous cross moods, and in the
relaxed condition incident to day-dreaming. Subject I also,
when nervously "on edge," indulges in excess movement.
It appears from the figures that increased slant and increased
horizontal extension are pretty closely associated, an asso-
ciation which one might anticipate on mechanical grounds
since the degree of inclination of the connecting stroke
would greatly influence the extent of territory covered by
the graphic product. A
question suggests itself as to the
possibility of untangling this mechanical relationship in
graphological interpretation Just at present, however, we

are not called upon to attempt such a feat. The record of
Subject II affords, however, some interesting examples of
slant and extension in mechanical opposition, since under
depression writing may be at once more compact and more
centrifugal than is usually the case.
If we turn to extreme individual records we find several
observations worthy of note. Thus the extreme up-align-
ment for Subject II, a rise of eight millimeters, is recorded
on November 8, just preceding a football game in which II

was intensely interested. The mood was characterized as
one of strain and excitement, certainly hyperkinetic, but not
describable as pleasurable or the reverse. On November
20, when the subject was ill with la grippe, the writing
shows a falling alignment, varying from two to thirteen
millimeters. The maximal slant (nearly fifty-three degrees)
occurs for Subject II on two days when the subject was in a
humorous mood, incident to editing the "yellow" number
of the college paper.
It is noticeable that often a given mood prevailed for a
number of days and continued to color graphic expression.
Thus, a number of II's extreme records fall in one and the
same week, a vacation period toned with a pleasant senti-
mental mood that magnified and inclined his writing. Four
of the contrasting records occur in one week and in the week
following five others; during this interval the subject was
struggling with la grippe.
number of Subject I's most diminutive specimens are
found in the first week of the experiment. The explanation
is obvious. There was definite concentration upon the
graphic product which resulted, as we have reason to ex-
pect, in small, even, somewhat vertical writing.
On the whole, the results of the experiment were most
enlightening. An increase in graphic movement accom-
panies heightened energy, while changes in slant and align-
ment appear influenced by emotional conditions but not in
in unequivocal way.

The is frequently made that graphic individu-
ality isbut a specific example of a pattern that is impressed
upon all the expressive movements of a given person. How
may one prove or disprove such an assertion? Obviously
not by casual observations which are subject to two very
serious sources of error: (i) the difficulty of accurately re-
porting on the individual character of expressive or graphic
patterns and (2) the biassing of observation in both cases by
a definite mental attitude which predisposes one to see simi-
larity or difference between the two.
As a control on such comparison it seems necessary that
the observations on the graphic and expressive pattern
should be made by different persons. It would be well if
each judge were ignorant of the specific point at issue, name-
ly, the extent to which the two sets of judgments would be
found to agree or disagree. In the simple test about to be
reported, this ignorance of the purpose of the experiment
existed only in the case of the observers of the expressive
movements. I myself passed judgment on the handwriting
My first attempt at handling the situation was unsuccess-
ful. After careful study of the handwriting of fourteen
students in one of my classes I attempted to describe the
graphic individuality of each by five carefully chosen de-
scriptive adjectives. From these adjectives I prepared a
list of words which I gave to the class, requesting each

member to choose three which should characterize as ac-
curately as possible the carriage, walk, and manner of ges-
ture of each member of the class. A
comparison of these
judgments with those I had passed on handwriting
showed cases of both agreement and disagreement.

I had, for example, characterized P's hand as expert,
graceful, mincing, rapid, and self-conscious. The following
tabulation was made of the judgments on his expressive
movements: Neat, 7; graceful, 3; mincing, 3; expert (in-
cluding fluent), 5; energetic, 2; conventional, 2; easy, 2;
diminutive, I decided, I affected, I smooth, 2 finished,
; ; ; ;

2; matter-of-fact, 2. The most frequently chosen adjective
is "neat" which might have been applied to P's writing with

great appropriateness. As his writing is small, "diminu-
tive" might also have been applied, etc. But, obviously, the
adjectives chosen were not sufficiently distinctive.
In at least one case, a very evident disagreement between
handwriting and type of movement is recorded. The ob-
servers agree fairly well on Pt. Her manner and walk are
characterized as decided, energetic, matter-of-fact, and
rapid. Her writing was characterized as neat, unaggressive,
unemphatic, and diminutive.
On the whole, however, the test proved of little value, not
only because of the vagueness of adjectives selected, but
also because of the inconclusiveness of the judgments passed
on the expressive movements. There are instances in which
twenty-one out of a possible thirty- four adjectives were ap-
plied to one and the same person.
Accordingly I planned a new experiment. In order to
control observations more decided to submit
definitely I

contrasting adjectives, with instructions to apply one of each
pair to the individual whose walk, carriage, and manner of
gesturing were under observation. Furthermore, I adopt-
ed as a general principle for choice of adjectives 'the dis-
tinctions that seem to hold for the contrasting types of ex-
plosive and inhibited writing. This resulted in a series of
paired adjectives as follows: Rapid or slow; light or heavy;
loose or compact; expansive or restrained; adroit or mala-
droit; fluent or jerky (tense); angular or supple (round-
ed); conventional or individual; impulsive or deliberate;
concentric or eccentric.

A blank record was prepared consisting of these paired
adjectives under the instructions, "Please study the walk,
:arriage, and gestures of the persons who are listed below
md thenclassify their usual movements under one of
<:ach of the following terms." The list of names given
was carefully selected and limited to twelve, as the passing
Df a real judgment demands considerable effort of attention.
With this in mind I also selected my collaborators with
bare. Six student judges were utilized, five of whom carried
DUt the exercise as part of their experimental work on the
general topic of expressive movement, in connection with
other experiments under the same general heading. Their
work was done very conscientiously. In addition, I asked
live faculty colleagues to pass judgment, selecting each for
pome definite reason; one, for example, was instructor in
My collaborators reported great difficulty in passing these
udgments. Most of them observed for some weeks the in-
rlividuals listed before they recorded their impressions. The
seemed particularly dif-
:lassification "concentric-eccentric"
icult to handle, possibly because the concept was a some-
what novel one. A note had, however, been appended to
he question blank, defining concentric as movement toward
lie body as a center eccentric as movement away from the

)ody. "Notice, for example, whether the elbows are car-
ied in or out, etc."Next to this division, that of "con-
was found most difficult to manage.
The terms "adroit" and "maladroit" proved ambiguous ;

ither grace or or expertness might be emphasized,
n some instances judges found themselves utterly unable
o reach a decision on some particular point for some par-
icular person observed, so that there are
inequalities in the
umber of judgments returned. Only seven judges return-
d records for P not because of any particular difficulty in
landling his case, but because of lack of acquaintance with
iim and failure in
opportunity to study his form of expres-

Before giving out my blanks for these records, my judg-
ments on the handwriting had been filed away. In some
cases I found great difficulty in reaching a decision and at
many points dissatisfaction with my record. I tried very

conscientiously to dismiss from my thoughts any character-
istics of an individual other than his handwriting, but it
would be impossible to assert that I succeeded absolutely in
such an endeavor. The rubrics which caused me the great-
est trouble were ''Light-Heavy" and "Individual-Conven-
tional." Note i. I had little confidence in either of these
sets of judgments except in a few extreme instances. My
judgment on the division "Fluent or Jerky (tense)" was in-
fluenced by a study of the line-quality under the microscope.
My observation on "Rapid-Slow" was based on general ap-
pearance. But it was possible to obtain, later, timed records
from all my subjects and to compare these records with an
order of merit arrangement earlier made by myself. The:
correlational coefficient was when the arrangement was
based on the normal writing and .61 when the correlation
was made with speeded writing. My biggest errors in
judgment was underrating the speed of D3's hand and over-
estimating that of P errors which were not confirmed byj
my collaborators' judgments on expression.
In passing judgment on concentric or eccentric move-!
ment I gave attention to slant. I grouped three specimens i

under the rubrics "concentric"; one was a backhand (L) a ;

second (Si) presented numerous examples of what the
French call "ecriture sinistrogyre," that is, curves or termi-
nal strokes turned in the reversed direction. S2 approaches
a vertical hand. 03 was classed as "eccentric" but aftei
considerable hesitation; this hand will, I believe, become
later a backhand, although as yet it follows the conventional
slant. 03 has since, in fact, informed me that in very rapic.
and careless writing there is a strong tendency to slanl

toward the left.

[Note i. I do not feel at all confident just what pair of term:
should be utilized in discriminating between the explosive and th'

ihibited hand with reference to this point. In the present set of
udgments I interpreted "individual" as equivalent to "easily identi-
ied," but such a definition causes an inclusion in such category of
.uids that are stylistic (Si) as well as those that are original. A
tylistic hand is, probably, inhibited. On the other hand, a careless
and (Wi), as such, shows explosive tendencies but it may not
jepart very
far from a conventional style.]

Table V gives a detailed survey of the results. The judg-
Inentson the expressive movements are summed under the
ippropriate heads while in the third column of each set
he graphological judgment is indicated by the initial letter
|f the chosen term. A
question mark after this initial indi-
ates uncertainty in decision; a plus mark shows that the
uality was evident to a high degree ; a minus sign indicates
As a rough approximation of the agreement between the
wo sets of judgments we may take the percentage of cases
p which the graphological judgment is in agreement with
he expressive judgment. Chance would account for a fifty
|er cent agreement; the actual agreement is 60.5 per cent,
nough higher than chance to point an interesting problem.
Jut such summary disposal of records is of much less in-
?rest than detailed perusal. For example, there are indi-
duals whose expressive movements are obviously character-
ed with ease as shown by the preponderance of judgments
i one direction or
another; there are others whose move-
ents are most As example of the first
difficult to classify.

pe we have Ai, Di, D2, P, Siof the second, D3, H, L,

V. Agreement of the expressive with the graphic judg-
A2, Di, H, P, Si less evident
ient is fairly consistent for ;

>r Bi, D2, 82, and W;
for Ai and D3 there is noteworthy
isagreement between the two sets of judgment; the judg-
ients on L are balanced. Agreement in the two sets of
idgments is, on the whole, more pronounced in case of the
icn, (A2, Di, B, Si, 82, and P) than of the women, (Ai,
)2, D3, H, L, and W). H
is an
outstanding exception to
iis statement. A greater degree of conventionality in
ther expression or handwriting on the part of women
bly explains this result.

uojspap ON

+ 1 M 1



(MONO rHOOTt<(Ni-l"*

+ +



O fl

1 S

l s

Age, even more than sex, cuts expressive tendencies. The
iyounger members of the group as such exhibit more impul-
isive and free movement than the older members. For D3
(and L
youth masks inhibitory tendencies very evident in
(theirhandwriting. But on the graphic side the effects are re-
ersed; age, with increasing graphic expertness, may lend

an impetuosity to handwriting movements which walk and
gesture lack. The record of B (one of the older group)
'should be studied in this connection. B's writing is im-
Ipetuous and excessively rapid in appearance. By the timed
records on normal writing it ranks second of the twelve.
;Note 2. But the general effect of carriage and walk is
slow and deliberate, although B's gestures are quick and
(impulsive. There is only one dissenting judgment in rating
as slow in movement ; that dissenting record is given by
'the instructor in dramatics who has had frequent occasion
to study B's movements in amateur theatricals.
Next to age, the confusion of certain bodily characteristics
Iwith movement seems to me a frequent source of conflict in
classification. Thus B is heavy of body but rather extra-
ordinarily light of movement if one discriminate with care;
D3, on the contrary, is excessively slight in build but rela-
tively heavy of step. The situation is further com-
plicated by the tendency to allow supposed mental traits to
influence thejudgment on expression. One subject for ob-
servation exhibits considerable inertia in getting down to
work but after beginning proceeds with great celerity and
with quick decision. This trait of inertia introduced diffi-
culty in classification of him as impulsive or deliberate.
If we turn from the individuals who were observed to the
rubrics employed we find that there is greatest agreement
n the two sets of judgments under the headings Rapid-Slow ;

Angular-Rounded; Impulsive-Deliberate; and Concentric-
Eccentric. Agreement is lesspronounced on Expansive-
Restrained; Fluent-Jerky; and Conventional-Individual.
[Note 2. In the speeded records B ties for first place. If speed
were accurately calculated on the basis of time per millimeter, he

would undoubtedly rank first by a safe margin, as his writing is
very large. B's speed could be increased by a reduction in size of
letters, a device which would certainly be adopted by a penman of
different mental type.]

There is slight preponderance in disagreement for the rub-
rics, Light-Heavy and Loose-Compact and only chance dis-
tribution for Adroit-Maladroit. Some disagreement is
probably to be attributed to ambiguity in the terms them-
selves. Thus I found myself interpreting "adroit" as equiv-
alent to "expert," "skillful," while I am inclined to think
that my collaborators stressed "grace." The records on con-
centric expression deserve particular study as the results
are most suggestive, and the agreement more extensive than
appears in a crude summing up by totals. Only in the case
of P is there striking disagreement. As my judgments had
in this case a rather definitely determined objective basis, the
results are enhanced in value.
In conclusion, be stated that the outcome of the
it may
experiment is slightly in favor of an agreement between
graphic and expressive movement but that the whole trend
of the results is indicative of the great difficulty inherent in

observation of expressive movement and the absence of all
standards for reference. Certainly no sweeping assertion of
general similarity can be ventured, although for a few traits
there is strong evidence of such harmony. It is rather in-

teresting to note in this connection that the percentage of
successes when judgments on character (see chap. VIII)
were related to particular graphic traits was usually higher
than that reported here. In the character investigation, I
had, however, the assistance of much more expert collabora-
tors than in the experiments on graphic individuality.

In 1906, Binet published his most interesting treatise on
"Les revelations de 1'ecriture d'apres un controle scien-
tifique," wherein he presented the results
of a series of care-
fully controlled tests designed to answer the following ques-
tions Does handwriting reveal sex, age, degree of intelli-

gence, character?
We have already had occasion to refer to Binet's con-
clusions with reference to revelation by handwriting of age
and sex. Here we may briefly summarize his conclusions
as to determination of intelligence and character from hand-

writing. Binet concludes that intelligence is revealed in
chirography although the extent of this revelation varies
with the individual the graphic signs of intelligence, grant-

ed an incontestable reality, are not always found in the writ-
ing of a man of great intelligence. In selecting from paired
specimens the hands produced by the more intelligent of the
penmen, Crepieux-Jamin gave 91 per cent, of successes.
(3c:101.) But the graphological portraits, correct so far as
they go, are often extremely vague. It is this vagueness
that needs to be cleared up by greater precision in definition
and interpretation of graphological signs. Perhaps the
conservative statement that there is more truth than error
in the judgments of graphologists anent intelligence sums up
Binet's records on this point. Concerning his tests on the
revelation of character in writing we may cite the general
conclusion that the errors in reading character from writing
are much greater than those found in reading intelligence
and that the graphologists show greater uncertainty in the
second than in the first test. In actual figures, Crepieux-
Jamin's percentage of successes was but 73 as compared
with 91 for intelligence (3c:248), where a chance success
of 50 might have been anticipated.


Binet's interest in the above investigations centered large-
ly in discovering a method for testing graphological con-
clusions. The question of method is, in fact, crucial. In
the experiment I wish to report here I have endeavored to
check up a chosen number of graphological principles by
utilizing a modified form of the order of merit method. In
many respects my procedure was very different from that
of Binet. In the first place I was obliged to dispense with
the services of professional graphologists. In lieu of their
interpretations I had recourse to measurements and observa-
tionson certain graphic signs that could be made by myself
and which could serve as a basis for a serial arrangement.
This arrangement the graphologi-
I shall refer to briefly as
cal arrangement. To
obtain a characterological arrange-
ment for correlation with the graphological one I was obliged
to ask help from a number of psychologists whose aver-
age judgment on a given individual I have taken as my
basis for comparison.
I may describe my procedure under four heads :
( i) choice
of graphic elements for measurement; (2) determination of
graphological scheme; (3) material to be used in the experi-
ment (4) questionary and collaborators.

(1) For graphic traits I was anxious to utilize as far as
it was
possible the same elements that I had studied in pre-
ceding sections of this book, namely size, slant, alignment,
continuity, line-quality or pressure, and proportion. These
graphic elements lent themselves in a greater or less degree
to objective measurement, so that it was possible on the basis
of such measurements to arrange a given collection of hands
into groups which represented a graded series in which a
given external character was present to a greater or less ex-
tent. As a matter of fact with reference to actual details
the procedure was not quite as simple as the above state-
ment would suggest. Modifications will become apparent in

the specific discussions that follow.
(2) Difficulties enough presented themselves when I tried
to correlate some simple graphological scheme with each

of the above mentioned elements. As finally worked out I
attempted a test of each of the following assumptions: I.
Small or filiform writing as an evidence of interest in min-
utiae or details in preference to preoccupation with princi-
ples; II. Large writing and, in particular, large capitals
in comparison with one-space letters as an indication of

pride, hopefulness III. Degree of discontinuity or discon-

nectedness of script as symptomatic of speculative or induc-
tive type of intellect in contrast to deductive or assimilative

type of thinking; IV. Pressure or line-quality and peculiar
forms of stroke as significant of aggressiveness; V. Varia-
tion in slant and alignment as symptomatic of temperament ;
VI. A complex of traits as symptomatic of an explosive ver-
sus an inhibited make-up.
(3) The was a series of letters from
collection utilized
thirty-six psychologists, most of whom were fairly well
knowm There are both advantages and disadvantages in-
herent in the choice of a closely selected group of individ-
uals. Such a close selection probably rules out extremes of
difference such as one might expect to find in a more miscel-
laneous group and so decreases the index of correlation. But
in testing the significance of a specific detail it seemed worth
while keeping fairly constant such factors as general culture,
general intelligence, and general character stability.
Several of my collaborators commented on the effect of
utilizing such a group of subjects. One remarks, "Is it not
to be expected that your list, which includes only those who
have a certain measure of success in psychology, would fall
in the medium class ?"
And another wrote as follows "In taking a group of well

known psychologists you have a group of successful people
which means in general a group whose members do not
differoutrageously from the social norm in social qualities
and energy and the qualities connected therewith deviate
in the direction of excess. Indeed as psychologists and as
successful they are a somewhat narrowly limited group."
Such a description is well worth keeping in mind.

(4) My collaborators in this experiment, whose group
judgment I utilized in getting a characterological rating,
consisted of twelve well-known psychologists who had had
considerable opportunity to know well many of the other
psychologists whose writing I was utilizing in the test. To
each of my collaborators I sent a list of thirty-six names
with the request that each individual be classified under
some one division of each of the given rubrics. The specific
groups utilized were as follows :

Group I. Preoccupation with Details in Contrast with Interest in Gen.
eral Principles
; Five Divisons.

1. Preo c c u-
pation with

crated to reduce a correlation that might have been obtained
under more favorable conditions.
My utilization of the graphological material was, of
course, somewhat mechanical. It is abundantly open to
criticism. In the first place I do not possess the background
that a professional graphologist would have and in the sec-
ond place I made little effort tomodify the outcome of me-
chanical measurement by what graphologists call the total
complex. In fact, I deliberately avoided any attempt to
utilize the graphological portrait as a whole and leaned
rather 'heavily upon details, with the very definite purpose
of testing as stringently as possible certain specific assump-
tions, with the assurance that positive results would thereby
be enhanced in value. Such procedure also protected me
from a possible inclination to be somewhat influenced by
the professional reputation of the penmen. personal My
acquaintance was limited to a few of the number. While
on the whole I rested my case largely on a few specific meas-
urements, there were times when I was obliged to have re-
course to a balancing of two or more graphic signs. Under
such circumstances my weighting of individual elements
was necessarily arbitrary as I could obtain little assistance
from the treatises on the subject.
Another prolific source of error was the possible multi-
plicity of causes for the same effect. Thus, increase in size
of writing may be determined by decrease in illumination or
increase in automatism or effort to mask a lack of motor
control. I had, of course, no knowledge of the conditions

under which the letters I was using had been produced and
no specific information concerning the penmen. In several
cases I should greatly like to know whether or not vision is
normal. Occasionally an age factor is slightly evident.
Probably the most significant source of error in the pres-
ent investigation was the lack in a number of instances of
sufficient material to justify a judgment. In the case of
two specimens I was certainly reckless in attempting to pass
judgment. One specimen (2) consisted of two lines scrawl-

ed with great precipitancy at the close of a typed letter and
the other (6) of a few words filling in the blanks on a print-
ed card. I have omitted my measurements on these speci-
mens from one or two groups as I shall specify later. Speci-
men 8 was also very meagre.
So was concerned, my collection was
far as letter-content
a fairly comparable one. Different size of letter sheets
probably introduced some slight uncertainty in judgment on
size but, on the whole, this source of error was negligible.
So much from the graphological side. A
few words con-
cerning the encountered by my collaborators in
the characterological judgment. Apart from the difficulty
inherent in breaking up such a closely selected group, there
was the error arising from insufficient acquaintance with an
individual or, at least, a more extensive acquaintance with
some of the psychologists than with others. As an outcome
of my returns I found it necessary to discard six of the
names included in my original list because of failure to re-
number of judgments to justify obtaining
ceive a sufficient
an average. Moreover, I did not receive for every group-
ing complete returns even for the twenty-nine remaining
names. My averages are obtained from individual items
varying from seven to twelve in number. Apparently my
collaborators found greatest difficulty in passing judgment
on Temperament, and on the Explosive-Inhibited make-up.
It moreover, evident that the categories I adopted
leave muchto be desired in the way of logical classifica-
tion. Such inadequacy was, in part at least, due to the dif-
ficulty I found in reducing graphological implications to any
form of system. Rigid definition of rubrics might have
been attempted, but I thought best in the present exploration
to employ somewhat elastic terms.
The attempt to work out a serial arrangement of hand-
writing specimens is something of a strain upon even the
most confident graphological principles. I have, therefore,
tabulated for each specimen under each rubric the graph-
ological group in which it was placed before the shaded ar-

rangement was made and I have given in parallel columns
the most common placement by my collaborators and the
mean variation from this mode. When the same number
of judgments was given for two adjacent groups, an inter-
mediate point between the two is indicated. When such
a distribution occurred for groups not adjacent, the miid-
group was used only when the shading from one to the other
was a quantitative one. Only a few cases of this sort
appear. See Table VI.
From a study of this table a number of conclusions be-
come evident which are not revealed by the citation of the
correlational coefficients. Such a tabulation enables us,
moreover, to pass from a mass treatment of results to an
estimation of the individual successes or failures that might
be anticipated from a graphological analysis. An individ-
ual, rather than statistical treatment, is, of course, the de-
sideratum in diagnostic tests.
i. Small or
filiform writing as an evidence of interest in
detail in contrast with speculative interests.
The correlation of an unusually small and precise hand
with love of minutiae and critical acumen, in contrast with
interest in far-reaching projects and speculative principles,
is a common one in the books on the
subjects. According-
ly I decided to classify collection of hands in five groups
to correspond with the characterological divisions previously

given: (i) Preoccupation with details; (2) Love of min-
utiae; critical, but wider interests than i; (3) Balanced at-
tention to details and principles; (4) Details subordinated;
interest in far-reaching projects; (5) Details ignored; spec-
ulative; poetic.
In making my classification for this purpose I had re-
course to actual measurement of size of letters. I was in-
terested in absolute size, not in proportional size, for ex-
ample the relative size of small and capital letters, a matter
of concern in another study. While small and narrow writ-
ing indicates love of detail, we have other graphic characters
that should be considered with it and these I utilized in

getting my group classification as follows: Regularity and
invariability of writing; even spacing of words and lines;
clear and sufficient formation of letters; care for punctua-
tion. Anorder of merit was prepared with these points in
mind. The contrasting hand involved the presence of one
or more of the following traits Rounded strokes extra

loops and high placed i-dot; rising alignment; insufficiently
clear letters and letters larger at the end than at the be-
ginning of a word.
In using absolute size of writing as a basis of classifica-
tion, one encounters the following difficulty, namely, that
size is a mbst variable aspect of writing it is influenced by

many factors, such as size of paper, the pen one uses, illu-
mination, physical condition, the care with which one writes.
Moreover, the group from which my collection came was
a closely selected one and did not show the range of varia-
tion in this respect that would be exhibited by a more
miscellaneous collection. The miniscules ranged in height
from less than .5 to 5 mm. My notes show the following
observations : "Almost all of these hands show attention to
detail in the clear-cut stroke and complete letter-form. There
are few 'speculative' hands among them." And again,
after completion of my arrangement, "I find it very difficult
to make an order of merit. Many of the hands are both
large and clear-cut. The final rating throws more
emphasis on even, complete, and careful writing than on
absolute size. Hands 7, 22, 9, 29 are taken as the central
group because they show to a high degree signs of both
critical acumen and interest in principles. The less evident
cases are thrown into groups 2 and 4. Groups I and 5 pre-
sent some interesting extremes. Except for groups I and
5, I have absolutely no confidence in this grouping."
But in spite of this lack of confidence the correlation
between the graphological order of merit and the one ob-
tained by averaging the group judgments of my collabor-
ators is high, .61 (P. E., .082). The most consistently
minute hands of the group seem written by psychologists

who concentrate by preference on details rather than prin-

ciples. There however, a few large hands in which
isize is overweighted by other symptoms of care for detail.

This outcomie of the experiment is of very great interest in
connection with our previous query whether small writing
such is evidence of general inhibitory tendencies. It indi-

cates an affirmative answer. Certainly the strongly hyper-
kinetic hands emanate from psychologists who are more

definitely interested in general theories than in detailed
criticism or prolonged experimentation. The blurring of let-
hand and the occasional trailing off
ters in the hyperkinetic
of the miniscules should not be confused with the even, clear-
Scut minuteness of the hand which is characterized above.
When we refer to Table VI, we find that there was com-
iplete agreement in the graphological and characterological
grouping fourteen times out of a possible twenty-nine
(including in this total the three half-step displacements).
The percentage of successful graphological placements is
J48.2 as against a possible 20 per cent, chance success. Further-
more, there are thirteen displacements of only one step. A
shift from Group i to Group 2 is less serious than one from

Group 2 to Group 3 or fromi Group 3 to Group 4. There
are two displacements, for specimens 16 and 27, which
between the graphological and
indicate decided disagreement
characterological judgment. Both are large, uneven, excit-
able hands. Probably both represent real contradiction of
the graphological position, although the very extreme varia-
tion on the characterological rating of 16 should be noted.
The placement of this psychologist by ten collaborators was
as follows: Group I (2) Group II (3) Group III (2)
; ; ;

Group IV (i) Group V (2).

2. Size and Emphasis of Capitals; Peeling of Self-
On the characterological side a five-fold grouping was
asked for: (i) Strong Feeling of Self- Worth (2) Moder-

ated Feeling of Self-Worth: (3) Average Feeling of Self-
Worth; (4) Modest Estimate of Self; (5) Excessive

Thegraphological arrangement was made on the basis of
size, emphasis, and ornamentation of capitals. In general,
such characteristics as increased size, emphasis, and orna-
mentation are explained as dependent upon the heightened
consciousness with which the capital is made. A
of a capital from the following small letter and its produc-
tion by a slow drawing movement is also evidence of aug-
mented consciousness. The increase in size with increase
in feeling of self- worth is explained ( I ) as a specific instance
of feeling for spatial relationship which correlates size with
prestige and (2) as a general outcome of effort, striving,
ambition. Subsidiary to size and emphasis of capitals comes
graphic size in general, involving small as well as large
In making my arrangement I first measured in millimeters
the capitals of each specimen and then measured the average
one-space letter; I also found the relative height of capital
and one-space letters. I then recorded observations on the

ornamentation, detachment, and design of the capital letters,
in this instance examining with special care the autograph
as in the autograph we find such features accentuated, and,
sometimes, striking variations introduced. In this connec-
tion it should be recalled that the utilization of relative height
of size of capital and one-space letter in a given collection of
specimens is of doubtful value in view of the fact that the
penmen may use writing systems in which the standard pro-
portion varies considerably. One who has been taught a
Spencerian system would for this very reason show a greater
difference between capital and one-space letters than one
who has learned a vertical system.
The correlation between the graphological arrangement
and the average character rating was inconclusive; .24 (P.
E., .12). There are some remarkable agreements, particu-
larly at the beginning and end of the two orders, but five very
great displacements. One of these big displacements occurs
for specimen 2, already cited as an inadequate representative.
Of the other four, two hands (8 and 4) exhibit, it would

seem, a deceptive appearance of ultra-modesty; while two
others (28 and 24) give, falsely, an impression of great self-
complacency. The characterological rating is, however, a
difficult one to handle. There are many different kinds of
self-feeling which manifest themselves in such different ways
as to make various impressions on one's acquaintances. For
example, pride sensitized by self-centeredness might impress
one less vigorously than a more candid and less self-con-
scious demeanor.
Table VI indicates that there were only six instances of
identical grouping by the two methods twelve cases where

a one-step displacement occurred; and twelve cases of still
greater displacement. The percentage of complete agreement
is no higher than mEght have been anticipated
by chance.
Actually, it is more significant 'than a chance agreement as
revealed by the graphological order of merit and the order
made on the basis of the average characterological rating.
To illustrate, No. 23 is first in the graphological order and
the characterological, ( (Group I (8) ; Group II (i) ) ;
first in
No. 26 is third in the graphological arrangement and second
in the characterological, ((Group I (5); Group 11(2));
No. 7 is fifth in the graphological order and third in the
characterological, ((Group I (6) Group II (3) Group III
; ;

(2)). The agreement between the graphological and char-
acteological rating is much more noticeable at the upper than
at -the lower end. Of the seven specimens placed in Group I
in the characterological column three were identified by a
graphological analysis but not one of those placed in Group
4 or 4.5 was selected by the graphological procedure. Cer-
tainly, there is nothing in specimens 10 or 20 that would lead
to expectation of less than average feeling of self- worth.
But a closer study of 6, on the basis of more material, has
convinced me of error in placing it.
Curiously, this group as a whole contains more individuals
who are characterized as possessing a strong feeling of
self-worth than those cited as excessively modest. A
professional graphologist might by utilization of the

whole graphological portrait have achieved greater success
than I was able to do in this limited application of a
principle obviously, I would have been in very considerable

error in about two-fifths of my readings. On the other hand,
I would have had some brilliant successes to balance my
failures !

3. Graphic Continuity; Originality of Mind versus
Poiver of Organisation.
From our previous discussion of graphic continuity it is
evident that the graphological interpretation of this element
is both ambiguous and uncertain.
Briefly, connected script
would seem to be the product of the practical organizing
type of mind broken script of the intuitive, original, fertile

type. No psychological reason for such interpretation is
attempted, and, as in every other case, multiplicity of causes
for the same effect is recognized.
In face of the somewhat bewildering disagreements in the
traditional literature I decided to confine myself to Preyer's
interpretation and with his five-fold scheme in mind (see
chapter iv) I turned to my collection of hands to attempt, if
possible,an arrangement into classes on the basis of degree
of connectedness or disconnectedness. I tabulated for each
specimen and for a constant number of words the number of
breaks between letters of the same word and also the number
of run-on words and the tying together by "t's." From this
tabulation I made my grouping. It was, however, necessary
to consider in tabulation of breaks that certain ones were
much more significant than others. For example, a break
between a capital letter and a following small one is less
significant than a break between the small letters of the
same words, largely because of the design of the capital and
its production by a separate impulse of attention.

My study of my collection revealed no sample of extreme
disconnection. It was possible, however, to make five groups,
as follows: (i) Numerous isolated letters and breaks in
letters;(2) Isolated groups of letters; (3) Occasional
breaks; (4) Highly connected hands, with (a) unattached

capitals; (b) attached capitals; (5) Completely connected
hands with words tied together. The margin of difference
between 'these groups was, however, very slight and after
consideration I threw together the last two groups. Per-
sonally I felt that the three hands thrown originally into
group 5 (namely, specimens 18, 22, and 23) represented a
very different form of motor impulse from that of group 4.
They engender a feeling of breathless precipitancy rather
than one of smooth expertness, and had I followed my per-
sonal feeling would have been grouped with i rather than 4.
Reverting now to Preyer, it was evident that his classifica-
tion could be applied to the present collection only in a modi-
fied form. His first class and possibly his second class were
not represented at all. My first division appeared to cor^es-
pond to his third ; my second division to his sub-group under
three. This latter division I eventually numbered I my first ;

group I re-numlbered 2 my third and fourth groups

remained as before. From the characterological side the
interpretation was as follows: (i) Original and fertile-
minded, little judgment or power of organization; (2) Orig-
inal and fertile-minded plus power of judgment; (3) Logical

type; combinative activity; (4) Assimilative capacity, utiliz-
ation of the ideas of others, neither critical nor ingenious.
As a scheme for logical classification the above is far from
satisfactory. As one of my collaborators wrote me in protest,
the divisions are not mutually exclusive. Personally I felt
the whole interpretation somewhat absurd, and my tabula-
tion on the basis of breaks highly fantastic.
Because of the small margin of differences it did not seem
feasible to miake an order of merit as I had in the preceding
cases. In preliminary comparison it seemed probable that
two groups would be sufficient to mark the real distinction
in degree of 'connectedness. This two- fold division could be
achieved by throwing together the first two and the last two
divisions. Those psychologists included in the first group
would be characterized by the predominance of originality;
those in the second by predominance of organizing, critical
and logical capacity.

Utilizing this twofold classification I found the returns
from my questionnaire interesting, and, in the light of my
scepticism, unexpected. From seven to eleven judgments
were passed upon 29 names of my list and in 20 of the 29
cases there was a preponderance of judgments in favor of
the graphological rating, 68.9 per cent of coincidence. For
example, specimen No. 27, graphologically in the second
group of the twofold division, is placed by five judges in
group 3, by two judges in group 4, and by one judge in
group i. Specimen No. 19 of the first graphological group
is placed by five judges in group I, by four in group 2, and

in group 4 by one judge. Nine of the judges consider the
penman original and fertile-minded rather than logical. Two
of the big discrepancies are found among the three specimens
that I placed at first in a fifth group by themselves and
included finally in group 4 with an inner protest.
This comparison of the graphological and characterologi-
crude to be satisfactory. We may, there-
cal rating is too
fore, turn to the tabulation of the fourfold grouping for
more complete returns. It will be observed that my
ators were very chary in utilization of the fourth group but
had found more use than I had for Group i. In fact I had
placed but one specimen in this group and, unfortunately,
received too few judgments on this psychologist to justify
using them. The tabulation indicates for the thirty hands
used a practical agreement on fourteen specimens, or 46.6
per cent of success as against a twenty-five per cent chance
agreement. There are twelve one-step displacements and
four greater displacements. A
shift from Group 2 to Group

3 or the reverse represents considerable error but a shift in
either direction for Groups i and 2 is not serious. The most
noticeable error is in placing 23, of which I have already
spoken. Probably in balancing all traits, as would be done
in a professional reading, this error would have been
The outcome of this part of the experiment is as interest-
ing as it was unexpected. It points to a problem which

should be investigated with care, namely, the possible cor-
relation of certain types of attention with both mental and
graphic traits. Of course, as always in this investigation,
we are confronted with a multiplicity of causes for the effect
under consideration, but in the present collection graphic
discontinuity can scarcely be attributed to inexpertness since
allthe penmen are experienced writers although not all are
expert penmen.
4. Line Quality; Aggressiveness.
Graphologists usually associate will-qualities with forceful
stroke or heavy pressure. As accessory signs they list angu-
and amplitude of writing, and the so-called
larity, verticality
dagger-stroke, evident in the terminal stroke or the bar of
the "t."
My fourth grouping was made on this basis to correspond
with the five divisions previously given, in a series graded
from great aggressiveness to passivity. The correlational
coefficient was too low to be significant, .23 (P. E., .13),
Two big displacements occur from one extreme in the
graphological arrangement to the other extreme in the char-
acterological, (Nos. 8 and 26). Omitting these two names
we get a correlation coefficient of .51 (P. E., .11), which
is high enough to have suggestive value. The omission of
the specimens indicated is, of course, illegitimate, for both
exhibit real contradictions to the graphological contention.
Both are unusually light tracings. We may, however, recall
in this connection that there a big margin of error in

attempting to estimate pressure by the eye.
The tabular summary of groups shows ten agreements
(inclusive of half-step displacements) or 33.3 per cent coin-
cidence as against a 20 per cent chance agreement. There
are thirteen one-step displacements and seven greater dis-
placements. Aone-place shift is not serious in this connec-
tion as the groups shade into one another. A character
reading on this basis would give about three successes out
of every four trials. The seven outstanding cases furnish
material for considerable analysis of details into which, how-
ever, it is not profitable to go at present. The greatest

discriminating between the merely
difficulty, I suspect, lies in
explosive hand and the hand which is both explosive and
aggressive. Probably a similar difficulty would be encoun-
tered in choosing the decisive character trait.
5. Slant and Alignment; Temperament.
The graphic elements that are correlated with emotional
and temperamental capacities include slant and alignment
on the ground that both are akin to movements of advance
or withdrawal as manifested in emotional expression in gen-
eral; the eccentric or centrifugal movement is correlated
with the pleasurable and the concentric or centripetal move-
ment with unpleasant feelings. The scheme utilized by the
graphologists would then be somewhat as follows :
of emotivity would be evidenced by degree of slant and by
its variability, while the direction which this emotivity would

take would be determined by the alignment; up-alignment,
optimism; down-alignment, depression; fluctuating align-
ment, variability. A few other signs of excitability and
variability might also be taken into consideration the pres-

ence, for example, of excess movements and general signs of
variability such as fluctuations in size. Moreover, large size
itself is sometimes cited as symptomatic of hopefulness ;

reduced dimensions, of depression. The characterological
grouping was as follows ( i ):
Optimistic, hopeful, enter-
prising; (2) Cheerful, active; (3) Equable, evenly active;
(4) Moody, variable, fluctuating attitude toward work;
(5) Pessimistic philosophy.
Preliminary to my arrangement of my collection I meas-
ured the average slant of each graphic specimen and esti-
mated the degree of slant variability. I then listed the
mannerisms of alignment, including not only the divergence
of the line itself from the horizontal but such alignment as
characterized the t-bar, since, according to Preyer, the manner
of crossing the "t" with an up or down or straight stroke is
but a special instance of alignment in general. I then selected
for my midgroup the equable in temperament the pro-
ducers of hands of slight and uniform slant, with straight

and uniform alignment and general evenness and calrrftiess
of script. This classification was not difficult except for
two specimens which combined minute size and vertical
slant, (3 and 22). I next sorted out the hands that showed

greater extremes and variability in slant, grading these with
respect both to degree of slant and of variability and separ-
ating into two groups on the basis of up or down alignment.
Sub-division of these groups was made on the basis of degree
of slant and of variability, but weight was also given to size
in separating Groups I and 2 and in Group 4 (Moody)
were included cases of fluctuating alignment, particularly the
line convex or concave in form. Group 5 included hands
with a perceptible down-alignment. I then arranged the

names within each group seriatim. Cases of fluctuating
alignment in combination with extreme slant were most
difficult to place and in a number of instances my notes indi-
cate a wavering in decision between Groups I and 4, with
uncertainty as to proper placement.
Slant and alignment as a basis of classification involve
great chance for error inasmuch as both are especially var-
iable. In making such observations as the above, one should
have at hand a number of specimens of a given hand, which
was not the case in this investigation except for a few pen-
men. In addition, there are the usual difficulties arising
from difference in system of writing utilized; a vertical
system encouraging a less degree of slant than the Spen-
cerian. It should, however, be observed that the slant in the
specimens of the present collection is extreme in only a very
few instances. A tendency to back-slant is perceivable in but
a few specimens and then only in spots. I have stated else-
where my conviction that such a tendency is related to latent
ambidextrality, a point which would have no connection
whatever with temperamental traits unless possibly the more
highly unidextrous person is more objective-minded and, in
consequence, more cheerful and hopeful in temperament
than the ambidextral type, an hypothesis highly speculative
but in harmony with a number of observations which I have

gathered. The present collection yielded only one specimen
of comiplete backhand writing and unfortunately too few of
my collaborators were acquainted with this individual to
make possible a characterization of him.
It was evident
that my collaborators found great difficulty
in arranging in temperamental groups the names sent them.
Considerable personal acquaintance is necessary before one
can confidently risk such a judgment. Just as I experienced
uncertainty in separating Groups I and 2, so too did certain
of my collaborators, indicating this by linking the two
groups. Although the groups do not give satisfactory basis
for an order of merit, I have attempted such a serial arrange-
ment as a possible way of bringing the records together.
The 'correlation between the graphological and character-
ological arrangement (.27, P. E., .12) is too low to be signi-
ficant. Study of the serial arrangements indicates, however,
an agreement on one-third of the twenty-seven names. A
big discrepancy occurs for specimen 19, placed graphologi-
cally in Group 4 but with a question mark, as possibly
belonging in I or 2. Seven of nine judges placed 19 in
either the first or second group so that the evidence of mood-
iness would seem deceptive. Specimen 4, placed by the
average judgment well up toward the cheerful end of the
spectrum}, exhibits, graphologically speaking, every sign of
a moody, fluctuating disposition. Five judges place 4 in
Group i one;
in Group 2 and three
in Group 4. Among
these latter judges is who probably knows 4
the individual
most intimately. Group 5 was very sparingly used by my
collaborators and the graphological arrangement included
only three names in this group. Only in one of these three
cases is the graphological rating in harmony with the aver-
age judgment. No. 17 is of especial interest in this connec-
tion. It presents an extraordinary fall in alignment, a
characteristic which I have noted in a number of specimens
of writing by this same penman. I have a feeling but with-
out specific information to back it that this hand presents
pathological features; it may be conditioned by defective

vision or general motor incoordination. Following the
graphological tradition I placed it in Group 5, but in this
case the falling alignment is not significant of a pessimistic
outlook as evidenced by the general agreement as to the
writer's cheerful or at least equable disposition.
The tabulation of the results by groups shows a complete
agreement on only six names, scarcely more than might be
expected by a chance arrangement. In two other cases the
results are indecisive. There are nine one-step displace-
ments ten displacements of more than one step. The tabu-

lation reveals in a number of instances very great discrep-
ancy in temperamental judgments as given by my collabor-
ators. No other arrangement gave so many extreme varia-
tions. For specimens 4, 5, 8, 18, and 26 the variation is so
extreme as to lead one to have as much confidence in the
graphological as in the personal rating.
6. Explosive and Inhibited Make-Up.
In testing the possibility of deducing from handwriting
explosive or inhibited tendencies, I turned from the more
conventional treatments in the usual treatises and adopted
Klages' scheme for determination of the degree to which
psychic energy is freely liberated or the reverse. The gen-
eral conception reminds us somewhat of James' description
of the obstructed and explosive types of will. We have here,
I suspect, the central problem in utilization of handwriting
in psychodiagnosis, just as we have many indications that
the distinction so well phrased by James is an essential one,
particularly if we recognize that an explosive type of will
may result either from defective inhibition or exaggerated
impulsion and an obstructed one from excessive inhibition
or insufficient impulsion.
Klages bases his scheme for identification of the free and
retarded hand on the results of experiments upon disguised
handwriting. The graphic characteristics that are accentu-
ated when control is at a maximum become then symlpto-
matic of a hand, the writer of which is highly self-conscious
and maintains a high degree of control without yielding to

automatism. On the other hand, the writer of the explosive
hand is one whose attention is directed away from the writ-

ing act, who confidently surrenders to graphic habits.
We have already seen that the maintenance of a high
degree of self-control shows itself in decreased size of writ-
ing, decreased slant, greater degree of disconnection, lessened
speed, increased pressure, and increased conventionality. Fol-
lowing Klages (26:152), therefore, I used one of each of
the following pairs of terms in description of each specimen :

1. Rapid or Slow.
2. Expansive or Restrained.
3. Pressure-weak or Pressure-strong.
4. Flowing or Intermittent.
a. Rounded or Angular.
b. Continuous or Broken.
5. Zealous or Retarded.
a. Open or Compact.
b. Inclined or Vertical.
6. Rich in Hxcess Movement or Meagre in Movement.
7. Centrifugal or Centripetal.
a. Right-slanted or Back-slanted.

b. Abductive or Adductive.
A. Emphasis of upper strokes or emphasis of
lower strokes.
B. Rising Alignment or Falling Alignment.
8. Assured Coordination or Unassured Coordination.
a. Slight or Excessive Difference in Lengths.

9. Individual or Stylistic.

In describing each specimen under such a scheme there
was, of course, considerable crossing over from one class to
another. I had difficulty also in determining with any degree
of accuracy the pressure of a hand and its speed. Experi-
mentalists warn us against an attempt to estimate force of
stroke from line quality, while from tests on myself I have
concluded that m(y judgment on the relative rapidity of a
hand is subject to considerable margin of error. Evidences

of extreme effort may blur effect of speed or a flowing move-
ment may enhance such appearance.
My study of my collection from the present point of view
proved most enlightening. It was undoubtedly possible to
pick out hands that gave evidence of explosive traits and
inhibited ones. Particularly was I interested in the question
whether the individual of strong impulses and exaggerated
inhibition could be discriminated from one with moderate
impulsion and deficient inhibition. I believe this to be pos-
sible. My collection contained no hands evidencing both
weak impulsion and deficient inhibition such individuals
achieve no measure of success but I have seen such char-
acterless hands in my experience with students.
my prelimjinary description I arranged the hands in
five groups: i. Excessive Impulsion; 2. Moderated Impul-
sion; 3. Balanced Impulsion; 4. Strong and Uneven Inhi-
bition 5. Excessive Inhibition. Then, as before, I attempted

to arrange my groups seriatim. This proved a baffling task
for I had, of course, no notion as to the proper method of
weighting the various graphic characters. As I have said
before some of my material was inadequate. In this connec-
tion I discarded two specimens (No. 2 and No. 6), the first

consisting of two lines sprawled in great haste at the close
of a type-written letter, the whole effect of which would be
to increase the signs of impulsion the second, consisting of

a few words filling in blanks on a card.
The correlation with the twenty-six other hands is 53.4,
a correlation high enough to suggest a most interesting and
promising field of work. With proper weighting and defini-
tion of graphic indications of impulsion and inhibition,
more extensive observation of graphic specimens, and
more adequate presentation of material a significant corre-
lation might be anticipated. It is along this line that I am

making exploration of a series of tests usable as an index
to temperamental patterns.
The two orders of merit indicate, it is true, certain cases
in which either my application of the scheme is at fault or

the graphological implications inaccurate. There are at least
two hands that give evidence both of extraordinary impul-
sion and of great conflict I have entered them as inhibited

types but my collaborators group them as explosive. There
is, on the contrary, one exceedingly smooth and supple hand

(10) that is produced by an individual of evidently inhibited
tendencies. Another specimen (8) baffles me completely.
Possibly I ami deceived by its extreme fluidity, the utter
absence of resistance to be overcome, in which case it may
characterize a person not only of great impulsion but also of
complete lack of conflicting or inhibiting tendencies.
Turning to the parallel tabulation of the most frequently
recorded character- judgment and the graphological group-
ing we find nine agreements in group-placement, or 32.1 per
cent of successes. There are eleven one-step displacements
and eight bad displacements. In the latter group fall hands
5, 7, 10, n, 15. 16, 24, and 30. Five of these errors occur
in connection with my group 5. These particular hands
bear the impress of very extreme graphic inhibition. I find
difficulty in believing that one would fail to find this par-
alleled in some form of temperamental inhibition, such as
undue reserve, timjidity, scrupulosity, extreme sensitiveness
and the like. But in the absence of evidence such an assump-
tion has, of course, no particular value.
On the psychological side we are pretty much in the
dark as to relation of impulse-tendencies and psychical types.
The returnsreceived in the present investigation suggested

the existence of some rather general character-patterns. I
therefore found it interesting to ask whether a comparison of
character judgments gave any evidence of what elements
constituted the explosive make-up. With this question in
mind I obtained the coefficients of correlation for the serial
arrangement on the basis of Bxplosiveness-Inhibition and
the other five characterological arrangements.
The results are unmistakable. The correlational coefficients
are all positive and high. It is evident that the explosive
type is characterized by a strong feeling of self-worth

(.81), and by aggressiveness (.82). To a less extent
the explosive type tends to be optimistic rather than pessimis-
tic (.45) and speculative rather than preoccupied with details

Certain individual divergencies are, however, of the ut-
most interest, largely because they may contribute to an
effort to distinguish between the explosive hand that is the
outcome of absence of inhibition and that which is explosive
in spite of inhibiting tendencies.

Light pressure is a sign of impulsion; heavy pressure of
inhibition, and, according to the traditional interpretation,
indicative of will-qualities. Certainly inhibitive tendencies
point to a mjore resistant make-up than the more fluidic
explosive type. Do our penmen of explosive type of writing,
inclusive of lightness of pressure, differ in any essential way
from those whose general type of hand is explosive but
heavy? Do the latter manifest dominant impulses breaking
forth from conflicting impulses? Is this type a more in-
elastic, dogmatic, self-critical pattern than the more fluid
The two penmen who gave such discrepant results when
we concerned ourselves with the arrangement for aggres-
siveness were of the light-pressure explosive pattern. Cer-
tainly the effect of their handwriting is radically different
from that of the heavy individualized hands that are at once
explosive and inhibited. Possibly the term "aggressive" is
not well-chosen in characterization of a dominant quality of
will ;
it possibly connote general impulsion to too high a
degree. In any case, our intercomparison of the character-
ological arrangements is of interest. There are two individ-
uals held to be somewhat more explosive than aggressive
and two others more aggressive than explosive. The first
two are also characterized as modest in their estimate of self,
giving us a pattern of non-aggressive, modest impetuosity;
the second two are thought to exhibit a more extreme feel-
ing of self- worth, giving a pattern of self-assured, aggressive
inhibition. These patterns appear to be much more excep-

tional than the aggressive prideful impulsive type or the

non-aggressive self-distrustful inhibited type. The tempera-
mental shows one inhibited individual who is,
none the less, Three explosive
of a cheerful cast of mind.
individuals are classified as variable in mood and fluctuating
in attitude. This raises a question which has hovered in the

background pretty persistently, namely, what application
should be mjade in this connection of the fluctuation in
explosive-inhibited tendencies so evident in unstable person-
alities ? The raising of this question must suffice at present.
A study of Table VI confirms the existence of certain
definite character patterns. Obviously 29, 24, 15, 12, n, 5,
and 2 belong to a balanced type. A
more extreme pattern
is suggested for 23, 26,
19, and 4. Again, a glance at the
table indicates great difference with respect to the certainty
with which the different individuals were grouped. There
is very extreme variation on 16, 17, 14, 19, and 22; much
less variation in the placing of 29, 23, 2, 10, and I. Of the
latter group the graphological rating of 29, 23, and I was
particularly successful, but the same cannot be said with
reference to 2 and 10. The difficulty with 2 was largely due
to an attempt to pass judgment on insufficient and hasty

writing; 10, however, as mentioned before, presents a real
problem for graphological analysis.
In general conclusion to this investigation it may be urged
that graphological contentions deserve more consideration
than they have received. Four of the six correlations between
graphological and charaeterological ratings that were put to
the test gave positive results, certainly much more striking
results than I should have ventured to anticipate. The
graphic deduction of feeling of self-
traits utilized for the
worth and the and alignment proved
utilization of slant

largely inconclusive, although even in these instances the
successes exceed those that might be dictated by chance. The
correlation of small, even, and clear-cut script with a critical
habit of mind; of a speculative tendency with broken script;
of aggressiveness with heavy line-quality and staccato

stroke; and of an explosive make-up with a hyperkinetic
hand should receive consideration. The instrumental study
of handwriting should find here certain problems worthy of
extensive investigation.
Klages' list of inhibited and explosive traits is largely
determined by features that characterize artificial or dis-
guised writing in contrast with spontaneous writing. But
it is interesting in this connection to
bring the results into
relationship with pathological writing. In general, the
observations are in harmony. Hyperkinetic writing, as char-
acterized by de Fursac, is exaggerated in size with excess of
flourishes and big lower loops it is a running hand, often

with words tied together it may be of such excessive speed

that certain letters are obliterated (effaced) ;
it is variable in
size, and slant, and presents malformations of the miniscules.
Its energy may be apparent either in the increased size, the
excessive rapidity, or the great pressure. Such a description
corresponds fairly well with Klages' more specific and
detailed scheme of graphic signs of release or checking of
impulsion. Variability is, however, more emphasized than
any particular kind of variation and pressure appears as an
element of hyperkinesis, rather than as a trait symptomatic
of retardation.
The hypokinetic or relaxed hand as it occurs in patho-
logical writing however, the parallel of the inhibited
is not,
hand, which exhibits, as it were, brakes put upon explosive-
ness. The signs of tension or inhibition listed by Klages are
the outcome of attention to writing, self-consciousness and
the like. They do not indicate disturbances of attention or
of motor coordination. The moderately inhibited hand is
more normal than the highly explosive.
It is possible to select hands from the present collection
that closely resemble those produced under psychic exalta-
tion. Specimen 26, for instance, shows such variations in
size from magnified capitals to miniscules that are only an
undulation of the pen as to closely resemble some of the
samples reproduced by de Fursac as characteristic of manic-

excitement. Two other hands of the collection show patho-
logical signs. One exhibits a fine tremor very evident under
the microscope and the other presents an extraordinary fall
in alignment.


J/V <> MCD * fO 10 t> IO -OO .CO-*IO OS lO'tfCOlOlO^Nt-rHlOt^lOrHC


,^,^,,,,v_, . . "*CO -OOCO i C0000510


A TT v
W 10 ^^~
* CO 00 10 T*H
00 CO IN rH t- IN t- IO CO OS







O O -OOlO^COCO O CO O t- 00 *O OS t- O
St-CNrH-* CO CO (N rH M <# 00 IO rH OS IO CO rH t- OS CO I- * .


| r^
We are now ready to venture upon a few concluding
remarks with reference to certain problems suggested in the
introductory chapter:
1 i ) The specific results of, and program for, investigation
precipitated by graphological discussions, and
(2) The possibility of utilizing graphic products in

diagnostic tests.
(i) To recapitulate briefly. We
have found reason to
believe that graphic size is symptomatic of the free release
of energy or the reverse and that extreme variation from
conventional standards has evidential value in an inter-group
comparison as well as in intra-individual comparison. We
have found that a high degree of variability in size, slant,
alignment, and similar graphic elements, witnesses lack of
mastery of the motor impulse by reason of defective control
or excessive impetuosity, and that there is reason to believe
that such extreme variability is evidence of the possession
of specific mental traits. Effortful control of graphic move-
ments likewise introduces very definite signs. haveWe
found reason to correlate frequent breaks in graphic con-
tinuity in an experienced hand with speculative interests, and
the contrasting hand with practicality. On the other hand,
we have encountered a stumbling block in our attempt to
and alignment in temperamental diagnosis. Even
utilize slant

here, however, indications were not wanting of some curious
confirmations of the graphological position. Such indication,
for example, is found in the correlative changes that
occurred with shifts in mood for Subject II in the investiga-
tion on individual variability, and in the curious similarity
between type of slant and expressive attitude which the
experiment on graphic individuality revealed in a number of
cases. The concept of the explosive versus the inhibited

jiand proved particularly enlightening although the limita-
ion of graphic expression by conventional standards greatly
,mbarrasses interpretation of symptoms.
Furthermore, the records from study of a given collection
If hands suggest that certain very specific psychodiagnostic
lorrelations deserve respectful consideration. Our percent-
ages of successes compared very favorably with those
obtained by Binet in his investigation in spite of the fact that
ve were dealing with specific rather than general correla-
ions, with a very limited range of material, and were
obliged to dispense with the services of the expert grapholo-
The detailed report of our employment of handwriting in
Dsychodiagnosis should be compared with Hollingsworth's
Investigation of the worth of judgments of character based
Upon study of a photograph. (24:41 f.). Limiting our-
'ielves to the records made by the individual judges in Hol-
ingworth's test (see page 52) we find that our one amateur
graphologist was, on the whole, rather more uniformly suc-
cessful. But the traits on which judgments were passed
Ivvere not, of course, directly comparable. It is rather inter-

esting in this connection to note that in the case of the one
trait where a comjparison may be instituted, namely, the

average estimation of "Conceit" from the photograph and
of "Feeling of Self-Worth" from handwriting, the second
source of information was slightly more accurate. All this
by the way.
Chiefly, our results are of value in that they outline a pro-
gram for further investigation. They witness the need of
more precise analysis of graphic elements and the influence
upon each of varying degrees of impulsion and of inhibition.
Another suggested problem is the relation of certain types
of attention to the smoothness or interruption of the motor
impulse. We have found some curious problems inherent
in back-slant ;
the causes that determine it should be investi-
gated. A
genetic study of the development of individuality
in hands should be undertaken, and, also, the
tracing of

similarity in the specific characteristics of family chirog-
(2) The attempt to utilize graphic products in diagnostic
tests involves, as a preliminary, scrutiny of possible classifi-
cation of mental types. A
number of simple bipartite classi-
fications which are current in psychological texts bear
obvious implication of a pattern fundamentally motor in
origin. The particular categories we have in mind have
been cited frequently in the preceding chapters. They include
the following organization of types Explosive or Ob-

structed Sensory or Motor Hyperkinetic or Hypokinetic
; ;

(akinetic). Another
distinction is rapidly becoming widely
accepted, namely, the division into an introverted or an
extroverted disposition. Although the angle of approach is
in this instance very different, the division itself effects a

very similar grouping of individuals and hence raises again
the question as to the relationship of motor impulses to these
psychic patterns, a question which suggests a method of
experimental attack of certain modern theories which up to
the present have been presented largely in dogmatic form.
With reference to the other organizations of types we
may say a few words. James' classic description of the
explosive and obstructed will has been appealed to in our
experimental sections. Certainly the varieties of reaction
characterized by him under the above terms have been found
most enlightening in our everyday comprehension of char-
acter, including as they do the two forms of explosive will
because of either exaggerated impulsion or defective inhibi-
tion, and the two forms of obstructed will, because of either
insufficient impulsion or excessive inhibition. Naturally one
expects to meet extreme types but rarely and one recognizes
the fact that emotional excitement may change the inhibited
individual into the explosive or that age may transform the
explosive person into the inhibited. Moreover, many indi-
viduals appear to fluctuate from one type to the other more
or less periodically.

Much that is said of the explosive-obstructed make-up is

very general in nature and not subject to experimental
analysis. Davenport's recent attempt to study the inherit-
ance of temperamental patterns and his conclusion that
defective inhibition of the nomadic instinct is "probably a
sex-linked, recessive, monohybrid trait" opens out the way
for a new method of study of mental types, although his
study of the more complex temiperamental patterns serves to
emphasize the obstacles that must be overcome.
The division into sensory-motor types was first made in
connection with reaction experiments. Now is not the time
to rehearse the varieties in interpretation of the outcome of
such experiments, nor the development of precision in
analysis. It would be venturesome in the extreme to attempt
to formulate any simple conclusions as to reaction-times and

temperamental patterns. We may, however, utilize the
terms sensory and motor in a purely descriptive way and
with Baldwin (2:163 f.) characterize the active or motile
person as very responsive to suggestion. "He tends to act
promptly, quickly, unreflectively generally such a person,
child or adult, is said to jump at conclusions Psychologically
such a person is dominated by Habit." He is domineering
and self-assertive; the man The "sensory child
of action.
is passive, more troubled by physical inertia, more contem-

plative when a little older, less apt in learning to act out new
movements, less quick at taking a hint." The sensory indi-
vidual is the observer, the thinker; he is non-suggestible,

non-expressive, non-self-revealing. Any simple registration
of such contrasting traits in a motor reaction would be of
greatest service in character analysis.
Hirt believes that there is a natural form of reaction deter-
mined by inborn constitution.
Moreover, his experiments
on writing have convinced him of a motor and sensorial
writing-type. The penmen who belong to the first type make
writing-movements those who belong to the second draw

"graphic signs." But is this motor or sensorial character of
writing a significant trait? And what proof have we that

differing psychophysical personalities are mirrored in these
reaction types? Hirt appears to assume the truth of this
latter proposition but he adds that an individual of one
may in a particular activity belong to a contrasting type. So
writing-type constitutes only one symptom out of many
possible symptoms. The motor reaction is the more rapid,

energetic, and emphatic; the sensorial more heavily slow
and controlled.
The third distinction to which we referred, namely, hyper-
kinetic or hypokinetic (akinetic) types is an outcome of study
of neuropathic or psychopathic constitutions. Of this distinc-
tion Southard (44) writes: "In confronting instances of
over- or under-activity, the analytical student should con-
sider in turn whether his given example of hyperkinesis is
hyperkinesis by defect or by excess and the same process is

of value in the analysis of akinetic phenomena."
Hirt observes that the manic and depressive make-ups
exhibit parallelism in psychic and in expressive activities
and lists self-confidence, indiscretion, mental energy, hasty,
unmotived and rapid acts as characteristic of the first retar-;

dation, indecision, anxiety, lack of self-confidence and inac-
tivity as characteristic of the second. From overt expression
one draws conclusions concerning the mental make-up.
In the study of temperamental organization, utilization of
some form of motor expression should, therefore, prove of
great value if it were possible to disentangle the charactero-
logical phases from those impressed upon the movement by
pressure of the environment. Handwriting suggests itself
as more convenient to utilize than manner of walk, gesture,
or emotional expression because of the fact that it produces
a record which can be utilized for repeated observation. It
is manifestly of great complexity and subject to great
environmental pressure, but in this respect it certainly pre-
sents no more difficulties than do other forms of expression.
And it would seem more simply susceptible of analysis than
is posture or walk or gesture.

Our preceding study has revealed distinctions in hands
(thoroughly in line with the theoretical categoriesies listed

above. But granting the existence of explosive and inhibited
(hands, of motor and sensory writing, of hyperkinetic and
hypokinetic chirography, what guarantee have we that they
point beyond themselves to a general motor make-up. Can
any specific motor pattern lead to inference of a general
temperamental pattern? Do not habit and training cause
strange inconsistencies in expression fluency, for example,

in speech but halting gesture ?

Signs of inhibition may indeed arise as an outcome of
motor conflict, but how variously such conflicts may be
motivated !
Reverting to inhibitions in graphic movements,
we have found that they may originate in shift from one
system of writing to a second, in bad eyesight, in transfer
from right to the left hand in writing or the reverse, or
even arise from the writing material that is utilized. On
the other hand, graphic smoothness or expertness or chiro-
graphic impetuosity rrtay possibly originate in ample prac-
tice, or in thorough grounding in the best form of graphic
movement. Does adequate comprehension of the multipli-
city of causes for graphic effects check any tendency to
diagnostic generalization? Yes, and no. It undoubtedly
enforces conservatism in attitude and insistence upon experi-
mental procedure but it does not place an impassable barrier
in the way of positive interpretation of results. The fore-
going account in its comparative treatment of the various
methods of studying handwriting and in its experimental
studies has furnished some indication of where to look for
positive results, sufficient material to at least encourage a
further search for a series of graphic tests which might be
utilized to get insight into the type of organization of a
reagent and so supplement intelligence tests. So far, of
course, as these tests concern the characterization of the
strength or weakness of the motor imfpulse, its energy or
free release or retardation they would have value in giving
the form of personality only. They would give us no

information as to the direction in which those impulses
would be applied, nor insight into the manifold individui
differences in fundamental impulses and sensitivities whicl
are basal to character organization.
Our further search along this line for diagnostic tests wil
consist not wholly in utilization of free handwriting but ir
such restrictions of it as arise in retarded, accelerated, dis-
guised, and automatic writing. In the hope of reaching
positive results I am now puttinr such tests to the proof.
i. Ayers, L. P. :
(a) A
Scale for Measuring the Quality of Han
writing of School Children. N. Y. Russ< :

Sage Foundation (1912).
(b) A
Scale for Measuring Quality of Ham
writing of Adults. N. Y. Russell Saj :

Foundation (1915).
2. Baldwin, J. M. The Story of the Mind. (1902).

3. Binet, A: (a) L'ecriture pendant les etats d'excitation. Ann
psychol., 9, (1903).
(b) Sur le grossissement provoque de 1'ecritu
Archiv de psychol., 4, (1904).
(c) Les revelations de 1'ecriture d'apres un contro
scientifique. (1906).
4. Bingham, W. V. : Some Norms of Dartmouth Freshman. Jot
Ed. Psy., 7, (1916).
5. Blackford, K. : Character Analysis.
-6. Booth, M. H. : How to Read Character in Handwriting.
Guide for the Beginner and Student in Graphc
ogy. (1910).
7. Brown W. : The Psychology of Writing. Rep. Brit. Ass. Ad
Sci. (1913).
8. Busse, H. H. Die Graphologie. (1895).

9. Crepieux Jamin, J. L'Ecriture et le Caractere. (1896).

10. Davenport, C. D. The Feebly Inhibited : Nomadism or tl :

Wandering Impulse with a Special Refe
ence to Heredity Inheritance and Temper;
ment. Carnegie Institute of Washingto
11. Dearborn, G. V. N. (a) Kinacsthesia and the Intelligent Wil

Amer. Jour. Psy., 24, (1913).
(b) Intuition. Psy. Rev. 22, (1917).
12. Deschamps, L. La philosophic de 1'ecriture. (1892.)

13. Diehl, A. Ueber die Eigenschaften der Schrift bei Gesunde

Psychol. Arbeiten, 3, (1899).
14 Downey, J. E. (a) Control Processes in Modified Handwri

ing. Psy. Rev. Monog. 37, (1908).
(b) Preliminary Study of Family Rescmblanc
in Handwriting. Uni. of Wyo. Dep't. (

Psy. Bull. No. i, (1910).
on the Sex of
^(c) Judgments Handwriting
Psy. Rev., 17, (1910).
(d) Judgments on Handwriting Similarity an
Difference. J. of Phil., Psy. and Sc
Meth., 11, (1914).
(e) Automatic Writing (with J. E. Anderson
Amer. Jour, of Psy., 26, (1915).

(f) Handwriting Disguise, Journal App. Psy.,
i, (1917)-
15. Erlenmeyer, A.: Die Schrift, Grundzuege ihrer Physiologic u.
Pathologic. (1879).
16. Freeman, F. N. :
(a) Preliminary Experiments on Writing Re-
actions. Psy. Rev. Monog., 34, (1907).
(b) Problems and Methods of Investigation
in Handwriting. Jour, of Ed. Psy., 3,
(c) Experimental Analysis of the Writing
Movement. Psy. Monog. 75, (1914).
(d) The Teaching of Handwriting. (1914).
17. French, W. L. : What Your Handwriting Reveals. (1914).
18. de Fursac, R. : Les Ecrits et les Dessins dans les Maladies ner-
(1905). veuses et mentales.
19. Gesell, A. L. Accuracy in Handwriting as related to School

Intelligence and Sex. Amer. Jour. Psy., 17,
20. Grommes, M. :Les Facteurs Musculaires du Graphism. Le Revue
Psy., 6, (1913)-
21 Gray, C. T. The Training of Judgment in the Use of the Ayers

Scale for Handwriting. Jour, of Ed. Psy,, 6,
22. Gross, A. :
Untersuchungen ueber die Schrift Gesunder und
Geisteskranker. Psy. Arb., 2, (1899).
23. Hirt, E. :
(a) Ueber empirisch begruendete Bewertung der nor-
malen und pathologischen Handschrift. Arch,
f. d. ges. Psy., 23, (1912).

(b) Untersuchungen ueber das Schreiben u. die
Schrift. Psy. Arb., 6, (1914).
24. Hollingrworth, H. L. Vocational Psychology. (1916). :

25. Holt, E. B. The Freudian Wish. (1916).

26. Klages, L. (a) Die Probleme der Graphologie Entwurf einer
: :

Psychodiagnostik. (1910).
(b) Begriff und Tatbestand der Handschrift,
Zsch. f. Psy., 63, (1912).
27 Koestcr, R. Die Schrift bei Geisteskrankheiten.
28. Lemaitre, A. Un Test base sur le modification de 1'ecriture.

Arch, de psy., 3, (1903).
29. Lomer, G. (a) Ein Fall von zirkularer Psychose, grapholo-

gisch gewuerdigt. Zsch. f. d. ges. Neur u. Psy-
chiat, 20, (1913).
(b) Ueber graphologische Kennzeichen des Schwach-
sinns. Arch. f. Psychiat, 53, (1914).
30. Mayer, M. Ueber die Beeinflussung der Schrift durch den Al-

kohol. Psy. Arb., 3, (1901).
31. McAllister, C. N. : Researches on Movements Used in Hand-
writing. Studies from Yale Psy. Lab., 8,

32. Meumann, Vorlesungen zur Einfuehrung in die Experimen-
E. :

telle Paedogogik und ihre psychologischen
Grundlagen, 2, (1907).
33. Meyer, A., Jelliffe, S. E., Hoch, A.: Dementia Praecox. (1911).
34. Meyer, G. (a) Ueber Schriftverstellung, Grap. Monatshefte,

4, (1900).
(b) Die wissenschaftlichen Grundlagen der Grapho-
logie. (1901).
35. Naecke, P. (a) Biologisches und Forensisches zur Handschrift.

Neur. Centbl., 30, (1911).
(b) Einige Bemerkungen bez. der Zeichnungen und
anderer Kuenstlerischer Aeusserungen von
Geisteskranken. Zsch. f. d. ges. Neur. u. Psy-
chiat, 17, (1913).
36. Osborn, A. S. (a) Questioned Documents.
: A
.Study of ques-
tioned documents with an outline of meth-
ods by which the facts may be discovered
and shown. (1910).
(b) Form Blindness or Psychology of Sight in
Relation to Legal Procedure. Case and
Comment, 19, (1913).
(c) Errors in Identification of Handwriting.
Amer. Law. Rev., 48, (1914).
37. Pellat, S.:L'Education aidee par la Graphologie. (1906).
38. Poppee, R. Graphologie. ( 1908)
: .

39. Preyer,W. Psychologic des Schreibens. (1895).

40. Prince, M. The Dissociation of a Personality. (1906).

41. Schneickert, H. (a) Graphologische
Uebergrif fe. Arch, f .

Krim.-Anthropol. u. Kriminalistik, 39,
(b) Die neue Handschriftensammlung der
Berliner Kriminalpolizei. Arch. f. Krim.-
Anthropol. u. Kriminalistik, 39, (1910).
42. Schneidemuehl, G. (a) Handschrift und Carakter. Ein Lehr-

resume of the researches and experiments
Grund wissenschaftlicher und prakti-
scher Studien. (1911).
(b) Die Psychologic der Handschrift.
Neur. Centbl., 31, (1912).
43 Starch, D. (a) Unconscious Imitation in Handwriting.
: Psy.
Rev., 18, (1911).
(b) The Similarity of Brothers and Sister in Men-
tal Traits. Psy. Rev., 24, (1917).
44. Southard, E. E. General Psychopathology. Psy. Bull. 14, (1917).

45. Thompson, M. E. Psychology and Pedagogy of Writing: a

resume of the researches and experiments
bearing on the history and pedagogy of
writing. (1911).

46. Thorndike, E. L. (a) Handwriting. Teachers College Record,

n, (1910).
(b) The resemblance of young twins in hand-
writing. Amer. Natural., 49, (1915).
47. Woodworth, R. S. The Accuracy of Voluntary Movement.

Psy. Rev. Monog. 13, (1899).

25 2